diff --git a/Data_base.csv b/Data_base.csv
new file mode 100644
index 0000000000000000000000000000000000000000..1e5e34cba3914d3c246f78550c74005a99905bd1
--- /dev/null
+++ b/Data_base.csv
@@ -0,0 +1,697 @@
+alloy,Fe,Ni,Co,Cr,V,Cu,VEC,AR1,AR2,PE,Density,TC,MP,FI,SI,TI,M,TEC,total ,range(oC),features,Fe.1,Ni.1,Co.1,Cr.1,V.1,Cu.1,Unnamed: 28,Unnamed: 29,Unnamed: 30,Unnamed: 31,Unnamed: 32,Unnamed: 33,Unnamed: 34,Unnamed: 35,Unnamed: 36,Unnamed: 37,Unnamed: 38,Unnamed: 39,Unnamed: 40,Unnamed: 41,Unnamed: 42
+Fe-Ni,1,0,0,0,0,0,8,140,124,1.83,7874,80,1181,762.47,1562.98,2957.4,2.22,11.92,1,30-100,VEC,8,10,9,6,5,11,,,,,,,,,,,,,,,
+,0.95,0.05,0,0,0,0,8.1,139.75,124.05,1.834,7925.7,80.55,1208.35,761.2035,1572.4825,2979.28,2.139,11.17,1,30-100,Atomic radius(empirical)(pm),140,135,135,140,135,135,,,,,,,,,,,,,,,
+,0.9,0.1,0,0,0,0,8.2,139.5,124.1,1.838,7977.4,81.1,1235.7,759.937,1581.985,3001.16,2.058,10.82,1,30-100,Atomic radius(Inoun)(pm),124,125,125,125,132,128,,,,,,,,,,,,,,,
+,0.85,0.15,0,0,0,0,8.3,139.25,124.15,1.842,8029.1,81.65,1263.05,758.6705,1591.4875,3023.04,1.977,9.91,1,30-100,Pauling electronegativity,1.83,1.91,1.88,1.66,1.63,1.9,,,,,,,,,,,,,,,
+,0.8,0.2,0,0,0,0,8.4,139,124.2,1.846,8080.8,82.2,1290.4,757.404,1600.99,3044.92,1.896,11.09,1,30-100,density(kg m-3),7874,8908,8900,7140,6110,8920,,,,,,,,,,,,,,,
+,0.75,0.25,0,0,0,0,8.5,138.75,124.25,1.85,8132.5,82.75,1317.75,756.1375,1610.4925,3066.8,1.815,13.6,1,30-100,Thermal conductivity(W m-1 K-1),80,91,100,94,30.7,400,,,,,,,,,,,,,,,
+,0.7,0.3,0,0,0,0,8.6,138.5,124.3,1.854,8184.2,83.3,1345.1,754.871,1619.995,3088.68,1.734,12.1,1,30-100,Melting point(K),1181,1728,1768,2180,2183,1357.77,,,,,,,,,,,,,,,
+,0.67,0.33,0,0,0,0,8.66,138.35,124.33,1.8564,8215.22,83.63,1361.51,754.1111,1625.6965,3101.808,1.6854,3.5,1,30-100,First ionisation energy(kJ mol-1),762.47,737.14,760.4,652.87,650.91,745.78,,,,,,,,,,,,,,,
+,0.64,0.36,0,0,0,0,8.72,138.2,124.36,1.8588,8246.24,83.96,1377.92,753.3512,1631.398,3114.936,1.6368,1.34,1,30-100,second ionisation energy(kJ mol-1),1562.98,1753.03,1648.39,1590.69,1412,1957.92,,,,,,,,,,,,,,,
+,0.635,0.365,0,0,0,0,8.73,138.175,124.365,1.8592,8251.41,84.015,1380.655,753.2246,1632.3483,3117.124,1.6287,1.3,1,30-100,third ionisation energy(kJ mol-1),2957.4,3395,3232.3,2987.1,2828.09,3554.6,,,,,,,,,,,,,,,
+,0.63,0.37,0,0,0,0,8.74,138.15,124.37,1.8596,8256.58,84.07,1383.39,753.0979,1633.2985,3119.312,1.6206,1.5,1,30-100,net magnetic moments/atom,2.22,0.6,1.72,-0.6,0,0,,,,,,,,,,,,,,,
+,0.6,0.4,0,0,0,0,8.8,138,124.4,1.862,8287.6,84.4,1399.8,752.338,1639,3132.44,1.572,4.07,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.55,0.45,0,0,0,0,8.9,137.75,124.45,1.866,8339.3,84.95,1427.15,751.0715,1648.5025,3154.32,1.491,7.11,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.5,0.5,0,0,0,0,9,137.5,124.5,1.87,8391,85.5,1454.5,749.805,1658.005,3176.2,1.41,9.68,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.4,0.6,0,0,0,0,9.2,137,124.6,1.878,8494.4,86.6,1509.2,747.272,1677.01,3219.96,1.248,11.28,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0.7,0,0,0,0,9.4,136.5,124.7,1.886,8597.8,87.7,1563.9,744.739,1696.015,3263.72,1.086,12.24,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.2,0.8,0,0,0,0,9.6,136,124.8,1.894,8701.2,88.8,1618.6,742.206,1715.02,3307.48,0.924,12.27,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.1,0.9,0,0,0,0,9.8,135.5,124.9,1.902,8804.6,89.9,1673.3,739.673,1734.025,3351.24,0.762,12.89,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,1,0,0,0,0,10,135,125,1.91,8908,91,1728,737.14,1753.03,3395,0.6,12.76,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+Fe-Co,0.91,0,0.09,0,0,0,8.09,139.55,124.09,1.8345,7966.34,81.8,1233.83,762.2837,1570.6669,2982.141,2.175,11.16,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.802,0,0.198,0,0,0,8.198,139.01,124.198,1.8399,8077.148,83.96,1297.226,762.0601,1579.8912,3011.8302,2.121,10.09,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.703,0,0.297,0,0,0,8.297,138.515,124.297,1.8449,8178.722,85.94,1355.339,761.8552,1588.3468,3039.0453,2.0715,9.71,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.654,0,0.346,0,0,0,8.346,138.27,124.346,1.8473,8228.996,86.92,1384.102,761.7538,1592.5319,3052.5154,2.047,9.75,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.604,0,0.396,0,0,0,8.396,138.02,124.396,1.8498,8280.296,87.92,1413.452,761.6503,1596.8024,3066.2604,2.022,9.58,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.506,0,0.494,0,0,0,8.494,137.53,124.494,1.8547,8380.844,89.88,1470.978,761.4474,1605.1725,3093.2006,1.973,9.33,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.407,0,0.593,0,0,0,8.593,137.035,124.593,1.8597,8482.418,91.86,1529.091,761.2425,1613.6281,3120.4157,1.9235,9.55,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.308,0,0.692,0,0,0,8.692,136.54,124.692,1.8646,8583.992,93.84,1587.204,761.0376,1622.0837,3147.6308,1.874,9.98,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.258,0,0.742,0,0,0,8.742,136.29,124.742,1.8671,8635.292,94.84,1616.554,760.9341,1626.3542,3161.3758,1.849,10.53,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.219,0,0.781,0,0,0,8.781,136.095,124.781,1.8691,8675.306,95.62,1639.447,760.8533,1629.6852,3172.0969,1.8295,11.45,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.209,0,0.791,0,0,0,8.791,136.045,124.791,1.8696,8685.566,95.82,1645.317,760.8326,1630.5393,3174.8459,1.8245,11.67,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.189,0,0.811,0,0,0,8.811,135.945,124.811,1.8706,8706.086,96.22,1657.057,760.7912,1632.2475,3180.3439,1.8145,11.8,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.159,0,0.841,0,0,0,8.841,135.795,124.841,1.8721,8736.866,96.82,1674.667,760.7291,1634.8098,3188.5909,1.7995,11.73,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.1,0,0.9,0,0,0,8.9,135.5,124.9,1.875,8797.4,98,1709.3,760.607,1639.849,3204.81,1.77,11.67,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.061,0,0.939,0,0,0,8.939,135.305,124.939,1.877,8837.414,98.78,1732.193,760.5263,1643.18,3215.5311,1.7505,11.8,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.041,0,0.959,0,0,0,8.959,135.205,124.959,1.878,8857.934,99.18,1743.933,760.4849,1644.8882,3221.0291,1.7405,12.52,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,0,1,0,0,0,9,135,125,1.88,8900,100,1768,760.4,1648.39,3232.3,1.72,12.13,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+Ni-Co,0,0.9,0.1,0,0,0,9.9,135,125,1.907,8907.2,91.9,1732,739.466,1742.566,3378.73,0.712,13.06,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,0.8,0.2,0,0,0,9.8,135,125,1.904,8906.4,92.8,1736,741.792,1732.102,3362.46,0.824,12.81,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,0.7,0.3,0,0,0,9.7,135,125,1.901,8905.6,93.7,1740,744.118,1721.638,3346.19,0.936,12.75,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,0.6,0.4,0,0,0,9.6,135,125,1.898,8904.8,94.6,1744,746.444,1711.174,3329.92,1.048,12.56,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,0.5,0.5,0,0,0,9.5,135,125,1.895,8904,95.5,1748,748.77,1700.71,3313.65,1.16,12.65,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,0.4,0.6,0,0,0,9.4,135,125,1.892,8903.2,96.4,1752,751.096,1690.246,3297.38,1.272,12.39,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,0.35,0.65,0,0,0,9.35,135,125,1.8905,8902.8,96.85,1754,752.259,1685.014,3289.245,1.328,12.41,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,0.3,0.7,0,0,0,9.3,135,125,1.889,8902.4,97.3,1756,753.422,1679.782,3281.11,1.384,12.08,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,0.25,0.75,0,0,0,9.25,135,125,1.8875,8902,97.75,1758,754.585,1674.55,3272.975,1.44,12.59,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,0.2,0.8,0,0,0,9.2,135,125,1.886,8901.6,98.2,1760,755.748,1669.318,3264.84,1.496,12.77,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,0.15,0.85,0,0,0,9.15,135,125,1.8845,8901.2,98.65,1762,756.911,1664.086,3256.705,1.552,12.48,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,0.1,0.9,0,0,0,9.1,135,125,1.883,8900.8,99.1,1764,758.074,1658.854,3248.57,1.608,12.61,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0,0.05,0.95,0,0,0,9.05,135,125,1.8815,8900.4,99.55,1766,759.237,1653.622,3240.435,1.664,12.32,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+FeNiCo,0.801,0.1,0.099,0,0,0,8.299,139.005,124.199,1.843,8078.974,83.08,1293.813,759.7321,1590.4406,3028.3751,2.0085,9.53,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.701,0.2,0.099,0,0,0,8.499,138.505,124.299,1.851,8182.374,84.18,1348.513,757.1991,1609.4456,3072.1351,1.8465,10.85,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.65,0.25,0.1,0,0,0,8.6,138.25,124.35,1.855,8235.1,84.75,1376.45,755.9305,1619.0335,3094.29,1.765,13.88,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.62,0.28,0.1,0,0,0,8.66,138.1,124.38,1.8574,8266.12,85.08,1392.86,755.1706,1624.735,3107.418,1.7164,4.38,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.601,0.3,0.099,0,0,0,8.699,138.005,124.399,1.859,8285.774,85.28,1403.213,754.6661,1628.4506,3115.8951,1.6845,7.59,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.58,0.32,0.1,0,0,0,8.74,137.9,124.42,1.8606,8307.48,85.52,1414.74,754.1574,1632.337,3124.922,1.6516,8.59,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.55,0.35,0.1,0,0,0,8.8,137.75,124.45,1.863,8338.5,85.85,1431.15,753.3975,1638.0385,3138.05,1.603,8.55,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.501,0.4,0.099,0,0,0,8.899,137.505,124.499,1.867,8389.174,86.38,1457.913,752.1331,1647.4556,3159.6551,1.5225,10.8,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.4,0.5,0.1,0,0,0,9.1,137,124.6,1.875,8493.6,87.5,1513.2,749.598,1666.546,3203.69,1.36,12.61,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0.6,0.1,0,0,0,9.3,136.5,124.7,1.883,8597,88.6,1567.9,747.065,1685.551,3247.45,1.198,7.72,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.2,0.7,0.1,0,0,0,9.5,136,124.8,1.891,8700.4,89.7,1622.6,744.532,1704.556,3291.21,1.036,6.89,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.1,0.8,0.1,0,0,0,9.7,135.5,124.9,1.899,8803.8,90.8,1677.3,741.999,1723.561,3334.97,0.874,7.48,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.701,0.1,0.199,0,0,0,8.399,138.505,124.299,1.848,8181.574,85.08,1352.513,759.5251,1598.9816,3055.8651,1.9585,8.5,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.602,0.2,0.198,0,0,0,8.598,138.01,124.398,1.8559,8283.948,86.16,1406.626,756.9941,1617.9012,3099.3502,1.797,9.88,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.551,0.25,0.199,0,0,0,8.699,137.755,124.449,1.86,8336.674,86.73,1434.563,755.7256,1627.4891,3121.5051,1.7155,10.98,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.522,0.28,0.198,0,0,0,8.758,137.61,124.478,1.8623,8366.668,87.04,1450.386,754.9677,1633.1052,3134.3582,1.6674,19.79,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.501,0.3,0.199,0,0,0,8.799,137.505,124.499,1.864,8388.374,87.28,1461.913,754.4591,1636.9916,3143.3851,1.6345,19.27,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.451,0.35,0.199,0,0,0,8.899,137.255,124.549,1.868,8440.074,87.83,1489.263,753.1926,1646.4941,3165.2651,1.5535,14.3,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.401,0.4,0.199,0,0,0,8.999,137.005,124.599,1.872,8491.774,88.38,1516.613,751.9261,1655.9966,3187.1451,1.4725,7.81,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.301,0.5,0.199,0,0,0,9.199,136.505,124.699,1.88,8595.174,89.48,1571.313,749.3931,1675.0016,3230.9051,1.3105,10.27,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.201,0.6,0.199,0,0,0,9.399,136.005,124.799,1.888,8698.574,90.58,1626.013,746.8601,1694.0066,3274.6651,1.1485,10.57,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.101,0.7,0.199,0,0,0,9.599,135.505,124.899,1.896,8801.974,91.68,1680.713,744.3271,1713.0116,3318.4251,0.9865,13.45,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.601,0.1,0.299,0,0,0,8.499,138.005,124.399,1.853,8284.174,87.08,1411.213,759.3181,1607.5226,3083.3551,1.9085,17.33,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.551,0.15,0.299,0,0,0,8.599,137.755,124.449,1.857,8335.874,87.63,1438.563,758.0516,1617.0251,3105.2351,1.8275,20.43,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.501,0.2,0.299,0,0,0,8.699,137.505,124.499,1.861,8387.574,88.18,1465.913,756.7851,1626.5276,3127.1151,1.7465,19.05,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.473,0.23,0.297,0,0,0,8.757,137.365,124.527,1.8633,8416.542,88.47,1481.149,756.0293,1632.0583,3139.6933,1.6989,20.25,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.451,0.25,0.299,0,0,0,8.799,137.255,124.549,1.865,8439.274,88.73,1493.263,755.5186,1636.0301,3148.9951,1.6655,19.78,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.401,0.3,0.299,0,0,0,8.899,137.005,124.599,1.869,8490.974,89.28,1520.613,754.2521,1645.5326,3170.8751,1.5845,11.6,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.301,0.4,0.299,0,0,0,9.099,136.505,124.699,1.877,8594.374,90.38,1575.313,751.7191,1664.5376,3214.6351,1.4225,13.69,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.201,0.5,0.299,0,0,0,9.299,136.005,124.799,1.885,8697.774,91.48,1630.013,749.1861,1683.5426,3258.3951,1.2605,13.09,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.101,0.6,0.299,0,0,0,9.499,135.505,124.899,1.893,8801.174,92.58,1684.713,746.6531,1702.5476,3302.1551,1.0985,12.11,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.501,0.1,0.399,0,0,0,8.599,137.505,124.499,1.858,8386.774,89.08,1469.913,759.1111,1616.0636,3110.8451,1.8585,13.81,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.454,0.15,0.396,0,0,0,8.696,137.27,124.546,1.8618,8435.396,89.57,1495.502,757.8508,1625.3099,3131.9004,1.779,13.58,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.424,0.18,0.396,0,0,0,8.756,137.12,124.576,1.8642,8466.416,89.9,1511.912,757.0909,1631.0114,3145.0284,1.7304,6.99,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.401,0.2,0.399,0,0,0,8.799,137.005,124.599,1.866,8490.174,90.18,1524.613,756.5781,1635.0686,3154.6051,1.6965,8.28,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.354,0.25,0.396,0,0,0,8.896,136.77,124.646,1.8698,8538.796,90.67,1550.202,755.3178,1644.3149,3175.6604,1.617,11.56,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.301,0.3,0.399,0,0,0,8.999,136.505,124.699,1.874,8593.574,91.28,1579.313,754.0451,1654.0736,3198.3651,1.5345,13.08,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.201,0.4,0.399,0,0,0,9.199,136.005,124.799,1.882,8696.974,92.38,1634.013,751.5121,1673.0786,3242.1251,1.3725,13.64,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.101,0.5,0.399,0,0,0,9.399,135.505,124.899,1.89,8800.374,93.48,1688.713,748.9791,1692.0836,3285.8851,1.2105,13.57,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.402,0.1,0.498,0,0,0,8.698,137.01,124.598,1.8629,8488.348,91.06,1528.026,758.9061,1624.5192,3138.0602,1.809,13.61,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.355,0.15,0.495,0,0,0,8.795,136.775,124.645,1.8668,8536.97,91.55,1553.615,757.6459,1633.7655,3159.1155,1.7295,12.04,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.302,0.2,0.498,0,0,0,8.898,136.51,124.698,1.8709,8591.748,92.16,1582.726,756.3731,1643.5242,3181.8202,1.647,5.44,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.202,0.3,0.498,0,0,0,9.098,136.01,124.798,1.8789,8695.148,93.26,1637.426,753.8401,1662.5292,3225.5802,1.485,9.7,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.102,0.4,0.498,0,0,0,9.298,135.51,124.898,1.8869,8798.548,94.36,1692.126,751.3071,1681.5342,3269.3402,1.323,11.6,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.302,0.1,0.598,0,0,0,8.798,136.51,124.698,1.8679,8590.948,93.06,1586.726,758.6991,1633.0602,3165.5502,1.759,13.9,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.256,0.15,0.594,0,0,0,8.894,136.28,124.744,1.8717,8638.544,93.53,1611.728,757.4409,1642.221,3186.3306,1.68,14.63,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.202,0.2,0.598,0,0,0,8.998,136.01,124.798,1.8759,8694.348,94.16,1641.426,756.1661,1652.0652,3209.3102,1.597,20.13,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.102,0.3,0.598,0,0,0,9.198,135.51,124.898,1.8839,8797.748,95.26,1696.126,753.6331,1671.0702,3253.0702,1.435,7.17,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.253,0.05,0.697,0,0,0,8.797,136.265,124.747,1.8689,8640.822,94.49,1617.489,759.7607,1632.0133,3170.8853,1.7905,5.02,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.203,0.1,0.697,0,0,0,8.897,136.015,124.797,1.8729,8692.522,95.04,1644.839,758.4942,1641.5158,3192.7653,1.7095,6.24,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.103,0.2,0.697,0,0,0,9.097,135.515,124.897,1.8809,8795.922,96.14,1699.539,755.9612,1660.5208,3236.5253,1.5475,7.52,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.031,0.22,0.749,0,0,0,9.189,135.155,124.969,1.8851,8869.954,97.4,1741.003,755.347,1668.7631,3259.5721,1.4891,12.06,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.011,0.24,0.749,0,0,0,9.229,135.055,124.989,1.8867,8890.634,97.62,1751.943,754.8404,1672.5641,3268.3241,1.4567,12.44,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.103,0.1,0.797,0,0,0,8.997,135.515,124.897,1.8779,8795.122,97.04,1703.539,758.2872,1650.0568,3220.2553,1.6595,11.85,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.041,0.16,0.799,0,0,0,9.119,135.205,124.959,1.8828,8859.214,97.74,1737.533,756.7633,1661.6306,3247.0611,1.5613,12.26,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.021,0.18,0.799,0,0,0,9.159,135.105,124.979,1.8844,8879.894,97.96,1748.473,756.2567,1665.4316,3255.8131,1.5289,12.81,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.051,0.1,0.849,0,0,0,9.049,135.255,124.949,1.8805,8848.474,98.08,1734.063,758.1796,1654.4981,3234.5501,1.6335,12.07,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.021,0.13,0.849,0,0,0,9.109,135.105,124.979,1.8829,8879.494,98.41,1750.473,757.4197,1660.1996,3247.6781,1.5849,12.72,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.051,0.05,0.899,0,0,0,8.999,135.255,124.949,1.879,8848.074,98.53,1736.063,759.3426,1649.2661,3226.4151,1.6895,11.98,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.031,0.07,0.899,0,0,0,9.039,135.155,124.969,1.8806,8868.754,98.75,1747.003,758.836,1653.0671,3235.1671,1.6571,12.55,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+,0.031,0.02,0.949,0,0,0,8.989,135.155,124.969,1.8791,8868.354,99.2,1749.003,759.999,1647.8351,3227.0321,1.7131,12.22,1,30-100,,,,,,,,,,,,,,,,,,,,,,
+Fe-Co-Cr,0.2,0,0.75,0.05,0,0,8.65,136.25,124.8,1.87,8606.8,95.7,1671.2,755.4375,1628.423,3165.06,1.704,10.65,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.25,0,0.7,0.05,0,0,8.6,136.5,124.75,1.8675,8555.5,94.7,1641.85,755.541,1624.1525,3151.315,1.729,9.72,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0,0.65,0.05,0,0,8.55,136.75,124.7,1.865,8504.2,93.7,1612.5,755.6445,1619.882,3137.57,1.754,9.61,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.35,0,0.6,0.05,0,0,8.5,137,124.65,1.8625,8452.9,92.7,1583.15,755.748,1615.6115,3123.825,1.779,9.56,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.4,0,0.55,0.05,0,0,8.45,137.25,124.6,1.86,8401.6,91.7,1553.8,755.8515,1611.341,3110.08,1.804,9.54,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.28,0,0.645,0.075,0,0,8.495,136.775,124.72,1.866,8480.72,93.95,1634.54,752.9149,1620.1477,3136.938,1.686,8.19,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0,0.625,0.075,0,0,8.475,136.875,124.7,1.865,8460.2,93.55,1622.8,752.9563,1618.4395,3131.44,1.696,7.39,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.32,0,0.605,0.075,0,0,8.455,136.975,124.68,1.864,8439.68,93.15,1611.06,752.9977,1616.7313,3125.942,1.706,7.34,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.34,0,0.585,0.075,0,0,8.435,137.075,124.66,1.863,8419.16,92.75,1599.32,753.0391,1615.0231,3120.444,1.716,7.5,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.36,0,0.565,0.075,0,0,8.415,137.175,124.64,1.862,8398.64,92.35,1587.58,753.0805,1613.3149,3114.946,1.726,8.8,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.38,0,0.545,0.075,0,0,8.395,137.275,124.62,1.861,8378.12,91.95,1575.84,753.1219,1611.6067,3109.448,1.736,9.23,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.405,0,0.52,0.075,0,0,8.37,137.4,124.595,1.8598,8352.47,91.45,1561.165,753.1736,1609.4715,3102.5755,1.7485,9.05,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.43,0,0.495,0.075,0,0,8.345,137.525,124.57,1.8585,8326.82,90.95,1546.49,753.2254,1607.3362,3095.703,1.761,9.4,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.345,0,0.575,0.08,0,0,8.415,137.125,124.655,1.8628,8405.23,92.62,1598.445,752.5118,1614.3076,3117.8435,1.7069,3.79,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.35,0,0.57,0.08,0,0,8.41,137.15,124.65,1.8625,8400.1,92.52,1595.51,752.5221,1613.8805,3116.469,1.7094,3.1,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.355,0,0.565,0.08,0,0,8.405,137.175,124.645,1.8623,8394.97,92.42,1592.575,752.5325,1613.4535,3115.0945,1.7119,3.38,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.315,0,0.6,0.085,0,0,8.43,137,124.685,1.8643,8427.21,93.19,1618.115,751.912,1616.5814,3124.8645,1.6803,6.47,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.325,0,0.59,0.085,0,0,8.42,137.05,124.675,1.8638,8416.95,92.99,1612.245,751.9327,1615.7273,3122.1155,1.6853,6.08,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.335,0,0.58,0.085,0,0,8.41,137.1,124.665,1.8633,8406.69,92.79,1606.375,751.9534,1614.8732,3119.3665,1.6903,4.05,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.345,0,0.57,0.085,0,0,8.4,137.15,124.655,1.8628,8396.43,92.59,1600.505,751.9741,1614.0191,3116.6175,1.6953,3.35,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.35,0,0.565,0.085,0,0,8.395,137.175,124.65,1.8625,8391.3,92.49,1597.57,751.9845,1613.592,3115.243,1.6978,3.07,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.355,0,0.56,0.085,0,0,8.39,137.2,124.645,1.8623,8386.17,92.39,1594.635,751.9948,1613.165,3113.8685,1.7003,3.3,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.365,0,0.55,0.085,0,0,8.38,137.25,124.635,1.8618,8375.91,92.19,1588.765,752.0155,1612.3109,3111.1195,1.7053,4.3,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.37,0,0.545,0.085,0,0,8.375,137.275,124.63,1.8615,8370.78,92.09,1585.83,752.0259,1611.8838,3109.745,1.7078,4.68,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.375,0,0.54,0.085,0,0,8.37,137.3,124.625,1.8613,8365.65,91.99,1582.895,752.0362,1611.4568,3108.3705,1.7103,6.88,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0,0,0.91,0.09,0,0,8.73,135.45,125,1.88,8741.6,99.46,1805.08,750.7223,1643.197,3210.232,1.5112,10.49,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.11,0,0.8,0.09,0,0,8.62,136,124.89,1.8745,8628.74,97.26,1740.51,750.95,1633.8019,3179.993,1.5662,10.06,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.21,0,0.7,0.09,0,0,8.52,136.5,124.79,1.8695,8526.14,95.26,1681.81,751.157,1625.2609,3152.503,1.6162,8.75,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.26,0,0.65,0.09,0,0,8.47,136.75,124.74,1.867,8474.84,94.26,1652.46,751.2605,1620.9904,3138.758,1.6412,7.14,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0,0.61,0.09,0,0,8.43,136.95,124.7,1.865,8433.8,93.46,1628.98,751.3433,1617.574,3127.762,1.6612,5.76,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.31,0,0.6,0.09,0,0,8.42,137,124.69,1.8645,8423.54,93.26,1623.11,751.364,1616.7199,3125.013,1.6662,5.36,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.32,0,0.59,0.09,0,0,8.41,137.05,124.68,1.864,8413.28,93.06,1617.24,751.3847,1615.8658,3122.264,1.6712,4.92,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.33,0,0.58,0.09,0,0,8.4,137.1,124.67,1.8635,8403.02,92.86,1611.37,751.4054,1615.0117,3119.515,1.6762,3.83,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.34,0,0.57,0.09,0,0,8.39,137.15,124.66,1.863,8392.76,92.66,1605.5,751.4261,1614.1576,3116.766,1.6812,3.16,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.35,0,0.56,0.09,0,0,8.38,137.2,124.65,1.8625,8382.5,92.46,1599.63,751.4468,1613.3035,3114.017,1.6862,2.69,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.355,0,0.555,0.09,0,0,8.375,137.225,124.645,1.8623,8377.37,92.36,1596.695,751.4572,1612.8765,3112.6425,1.6887,1.52,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.36,0,0.55,0.09,0,0,8.37,137.25,124.64,1.862,8372.24,92.26,1593.76,751.4675,1612.4494,3111.268,1.6912,0.91,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.365,0,0.545,0.09,0,0,8.365,137.275,124.635,1.8618,8367.11,92.16,1590.825,751.4779,1612.0224,3109.8935,1.6937,0.22,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.37,0,0.54,0.09,0,0,8.36,137.3,124.63,1.8615,8361.98,92.06,1587.89,751.4882,1611.5953,3108.519,1.6962,-1.07,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+(?),0.38,0,0.53,0.09,0,0,8.35,137.35,124.62,1.861,8351.72,91.86,1582.02,751.5089,1610.7412,3105.77,1.7012,4.05,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.415,0,0.495,0.09,0,0,8.315,137.525,124.585,1.8593,8315.81,91.16,1561.475,751.5814,1607.7519,3096.1485,1.7187,8.29,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.16,0,0.745,0.095,0,0,8.555,136.275,124.84,1.872,8568.64,96.23,1713.22,750.5159,1629.2429,3165.022,1.5796,9.8,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.21,0,0.695,0.095,0,0,8.505,136.525,124.79,1.8695,8517.34,95.23,1683.87,750.6194,1624.9724,3151.277,1.6046,8.28,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.26,0,0.645,0.095,0,0,8.455,136.775,124.74,1.867,8466.04,94.23,1654.52,750.7229,1620.7019,3137.532,1.6296,7.02,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.29,0,0.615,0.095,0,0,8.425,136.925,124.71,1.8655,8435.26,93.63,1636.91,750.785,1618.1396,3129.285,1.6446,5.92,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.31,0,0.595,0.095,0,0,8.405,137.025,124.69,1.8645,8414.74,93.23,1625.17,750.8264,1616.4314,3123.787,1.6546,5.03,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.33,0,0.575,0.095,0,0,8.385,137.125,124.67,1.8635,8394.22,92.83,1613.43,750.8678,1614.7232,3118.289,1.6646,3.98,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.34,0,0.565,0.095,0,0,8.375,137.175,124.66,1.863,8383.96,92.63,1607.56,750.8885,1613.8691,3115.54,1.6696,3.22,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.345,0,0.56,0.095,0,0,8.37,137.2,124.655,1.8628,8378.83,92.53,1604.625,750.8988,1613.4421,3114.1655,1.6721,2.3,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.355,0,0.55,0.095,0,0,8.36,137.25,124.645,1.8623,8368.57,92.33,1598.755,750.9195,1612.588,3111.4165,1.6771,0.78,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.36,0,0.545,0.095,0,0,8.355,137.275,124.64,1.862,8363.44,92.23,1595.82,750.9299,1612.1609,3110.042,1.6796,0.55,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.365,0,0.54,0.095,0,0,8.35,137.3,124.635,1.8618,8358.31,92.13,1592.885,750.9402,1611.7339,3108.6675,1.6821,0.12,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.37,0,0.535,0.095,0,0,8.345,137.325,124.63,1.8615,8353.18,92.03,1589.95,750.9506,1611.3068,3107.293,1.6846,-0.6,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.375,0,0.53,0.095,0,0,8.34,137.35,124.625,1.8613,8348.05,91.93,1587.015,750.9609,1610.8798,3105.9185,1.6871,-0.06,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+(?)(?),0.38,0,0.525,0.095,0,0,8.335,137.375,124.62,1.861,8342.92,91.83,1584.08,750.9713,1610.4527,3104.544,1.6896,7.31,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0,0.6,0.1,0,0,8.4,137,124.7,1.865,8416.2,93.4,1633.1,750.268,1616.997,3125.31,1.638,5.14,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.31,0,0.59,0.1,0,0,8.39,137.05,124.69,1.8645,8405.94,93.2,1627.23,750.2887,1616.1429,3122.561,1.643,4.52,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.33,0,0.57,0.1,0,0,8.37,137.15,124.67,1.8635,8385.42,92.8,1615.49,750.3301,1614.4347,3117.063,1.653,3.76,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.35,0,0.55,0.1,0,0,8.35,137.25,124.65,1.8625,8364.9,92.4,1603.75,750.3715,1612.7265,3111.565,1.663,1.4,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.355,0,0.545,0.1,0,0,8.345,137.275,124.645,1.8623,8359.77,92.3,1600.815,750.3819,1612.2995,3110.1905,1.6655,0.74,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.36,0,0.54,0.1,0,0,8.34,137.3,124.64,1.862,8354.64,92.2,1597.88,750.3922,1611.8724,3108.816,1.668,0.2,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.365,0,0.535,0.1,0,0,8.335,137.325,124.635,1.8618,8349.51,92.1,1594.945,750.4026,1611.4454,3107.4415,1.6705,-0.14,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.37,0,0.53,0.1,0,0,8.33,137.35,124.63,1.8615,8344.38,92,1592.01,750.4129,1611.0183,3106.067,1.673,0.18,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+(?)(?),0.375,0,0.525,0.1,0,0,8.325,137.375,124.625,1.8613,8339.25,91.9,1589.075,750.4233,1610.5913,3104.6925,1.6755,2.87,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0,0.595,0.105,0,0,8.385,137.025,124.7,1.865,8407.4,93.37,1635.16,749.7304,1616.7085,3124.084,1.6264,5.8,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.31,0,0.585,0.105,0,0,8.375,137.075,124.69,1.8645,8397.14,93.17,1629.29,749.7511,1615.8544,3121.335,1.6314,5.33,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.32,0,0.575,0.105,0,0,8.365,137.125,124.68,1.864,8386.88,92.97,1623.42,749.7718,1615.0003,3118.586,1.6364,4.53,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.345,0,0.55,0.105,0,0,8.34,137.25,124.655,1.8628,8361.23,92.47,1608.745,749.8235,1612.8651,3111.7135,1.6489,2.06,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.35,0,0.545,0.105,0,0,8.335,137.275,124.65,1.8625,8356.1,92.37,1605.81,749.8339,1612.438,3110.339,1.6514,1.66,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+(?*),0.37,0,0.525,0.105,0,0,8.315,137.375,124.63,1.8615,8335.58,91.97,1594.07,749.8753,1610.7298,3104.841,1.6614,8.73,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.38,0,0.515,0.105,0,0,8.305,137.425,124.62,1.861,8325.32,91.77,1588.2,749.896,1609.8757,3102.092,1.6664,10.73,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.355,0,0.535,0.11,0,0,8.315,137.325,124.645,1.8623,8342.17,92.24,1604.935,749.3066,1611.7225,3107.7385,1.6423,11.97,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.375,0,0.515,0.11,0,0,8.295,137.425,124.625,1.8613,8321.65,91.84,1593.195,749.348,1610.0143,3102.2405,1.6523,11.04,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.21,0,0.675,0.115,0,0,8.445,136.625,124.79,1.8695,8482.14,95.11,1692.11,748.4688,1623.8184,3146.373,1.5582,8.65,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.26,0,0.625,0.115,0,0,8.395,136.875,124.74,1.867,8430.84,94.11,1662.76,748.5723,1619.5479,3132.628,1.5832,7.23,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0,0.585,0.115,0,0,8.355,137.075,124.7,1.865,8389.8,93.31,1639.28,748.6551,1616.1315,3121.632,1.6032,6.04,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.32,0,0.565,0.115,0,0,8.335,137.175,124.68,1.864,8369.28,92.91,1627.54,748.6965,1614.4233,3116.134,1.6132,10.24,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.34,0,0.545,0.115,0,0,8.315,137.275,124.66,1.863,8348.76,92.51,1615.8,748.7379,1612.7151,3110.636,1.6232,12.21,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.39,0,0.495,0.115,0,0,8.265,137.525,124.61,1.8605,8297.46,91.51,1586.45,748.8414,1608.4446,3096.891,1.6482,13.39,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.36,0,0.52,0.12,0,0,8.28,137.4,124.64,1.862,8319.44,92.08,1606.12,748.2416,1610.7184,3103.912,1.6216,13.09,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.1,0,0.75,0.15,0,0,8.45,136.25,124.9,1.875,8533.4,97.1,1771.1,744.4775,1631.194,3168.03,1.422,10.25,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.15,0,0.7,0.15,0,0,8.4,136.5,124.85,1.8725,8482.1,96.1,1741.75,744.581,1626.9235,3154.285,1.447,10.23,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.2,0,0.65,0.15,0,0,8.35,136.75,124.8,1.87,8430.8,95.1,1712.4,744.6845,1622.653,3140.54,1.472,9.51,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.225,0,0.625,0.15,0,0,8.325,136.875,124.775,1.8688,8405.15,94.6,1697.725,744.7363,1620.5178,3133.6675,1.4845,12.28,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.25,0,0.6,0.15,0,0,8.3,137,124.75,1.8675,8379.5,94.1,1683.05,744.788,1618.3825,3126.795,1.497,13.97,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0,0.55,0.15,0,0,8.25,137.25,124.7,1.865,8328.2,93.1,1653.7,744.8915,1614.112,3113.05,1.522,15.38,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.35,0,0.5,0.15,0,0,8.2,137.5,124.65,1.8625,8276.9,92.1,1624.35,744.995,1609.8415,3099.305,1.547,15.98,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0,0,0.8,0.2,0,0,8.4,136,125,1.88,8548,98.8,1850.4,738.894,1636.85,3183.26,1.256,10.84,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.1,0,0.7,0.2,0,0,8.3,136.5,124.9,1.875,8445.4,96.8,1791.7,739.101,1628.309,3155.77,1.306,10.48,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.15,0,0.65,0.2,0,0,8.25,136.75,124.85,1.8725,8394.1,95.8,1762.35,739.2045,1624.0385,3142.025,1.331,13.33,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.2,0,0.6,0.2,0,0,8.2,137,124.8,1.87,8342.8,94.8,1733,739.308,1619.768,3128.28,1.356,14.78,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.25,0,0.55,0.2,0,0,8.15,137.25,124.75,1.8675,8291.5,93.8,1703.65,739.4115,1615.4975,3114.535,1.381,15.65,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0,0.5,0.2,0,0,8.1,137.5,124.7,1.865,8240.2,92.8,1674.3,739.515,1611.227,3100.79,1.406,16.31,1,20-60,,,,,,,,,,,,,,,,,,,,,,
+Fe-Co-Cr-Cu,0.45,0,0.5,0,0,0.05,8.65,137.25,124.7,1.8585,8439.3,106,1483.35,760.6005,1625.432,3124.71,1.859,10.49,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.425,0,0.525,0,0,0.05,8.675,137.125,124.725,1.8598,8464.95,106.5,1498.025,760.5488,1627.5673,3131.5825,1.8465,10.18,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.4,0,0.55,0,0,0.05,8.7,137,124.75,1.861,8490.6,107,1512.7,760.497,1629.7025,3138.455,1.834,10.17,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.25,0,0.7,0,0,0.05,8.85,136.25,124.9,1.8685,8644.5,110,1600.75,760.1865,1642.514,3179.69,1.759,11.59,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.53,0,0.4,0.02,0,0.05,8.51,137.75,124.62,1.8545,8322.02,104.28,1444.63,758.6155,1617.4452,3097.814,1.8526,10.09,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.48,0,0.45,0.02,0,0.05,8.56,137.5,124.67,1.857,8373.32,105.28,1473.98,758.512,1621.7157,3111.559,1.8276,9.71,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.455,0,0.475,0.02,0,0.05,8.585,137.375,124.695,1.8583,8398.97,105.78,1488.655,758.4603,1623.851,3118.4315,1.8151,10,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.43,0,0.5,0.02,0,0.05,8.61,137.25,124.72,1.8595,8424.62,106.28,1503.33,758.4085,1625.9862,3125.304,1.8026,10.12,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.405,0,0.525,0.02,0,0.05,8.635,137.125,124.745,1.8608,8450.27,106.78,1518.005,758.3568,1628.1215,3132.1765,1.7901,10.3,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.38,0,0.55,0.02,0,0.05,8.66,137,124.77,1.862,8475.92,107.28,1532.68,758.305,1630.2567,3139.049,1.7776,10.19,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.33,0,0.6,0.02,0,0.05,8.71,136.75,124.82,1.8645,8527.22,108.28,1562.03,758.2015,1634.5272,3152.794,1.7526,10.51,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.45,0,0.45,0.05,0,0.05,8.5,137.5,124.7,1.8585,8351.3,105.7,1503.95,755.224,1622.547,3112.45,1.743,10.32,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.4,0,0.5,0.05,0,0.05,8.55,137.25,124.75,1.861,8402.6,106.7,1533.3,755.1205,1626.8175,3126.195,1.718,10.3,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.35,0,0.55,0.05,0,0.05,8.6,137,124.8,1.8635,8453.9,107.7,1562.65,755.017,1631.088,3139.94,1.693,10.86,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.25,0,0.65,0.05,0,0.05,8.7,136.5,124.9,1.8685,8556.5,109.7,1621.35,754.81,1639.629,3167.43,1.643,11.62,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.35,0,0.525,0.075,0,0.05,8.525,137.125,124.8,1.8635,8409.9,107.55,1572.95,752.3288,1629.6455,3133.81,1.635,9.93,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.33,0,0.525,0.095,0,0.05,8.485,137.125,124.82,1.8645,8395.22,107.83,1592.93,750.1368,1630.1997,3134.404,1.5786,5.87,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.45,0,0.4,0.1,0,0.05,8.35,137.75,124.7,1.8585,8263.3,105.4,1524.55,749.8475,1619.662,3100.19,1.627,10.5,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.4,0,0.45,0.1,0,0.05,8.4,137.5,124.75,1.861,8314.6,106.4,1553.9,749.744,1623.9325,3113.935,1.602,10.6,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.35,0,0.5,0.1,0,0.05,8.45,137.25,124.8,1.8635,8365.9,107.4,1583.25,749.6405,1628.203,3127.68,1.577,6.4,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.325,0,0.525,0.1,0,0.05,8.475,137.125,124.825,1.8648,8391.55,107.9,1597.925,749.5888,1630.3383,3134.5525,1.5645,5.55,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0,0.55,0.1,0,0.05,8.5,137,124.85,1.866,8417.2,108.4,1612.6,749.537,1632.4735,3141.425,1.552,6.38,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.29,0,0.56,0.1,0,0.05,8.51,136.95,124.86,1.8665,8427.46,108.6,1618.47,749.5163,1633.3276,3144.174,1.547,7.6,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.25,0,0.6,0.1,0,0.05,8.55,136.75,124.9,1.8685,8468.5,109.4,1641.95,749.4335,1636.744,3155.17,1.527,10.23,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.315,0,0.525,0.11,0,0.05,8.455,137.125,124.835,1.8653,8384.21,108.04,1607.915,748.4928,1630.6154,3134.8495,1.5363,7.23,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.375,0,0.45,0.125,0,0.05,8.35,137.5,124.775,1.8623,8296.25,106.75,1578.875,747.004,1624.6253,3114.6775,1.5315,16.72,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.325,0,0.5,0.125,0,0.05,8.4,137.25,124.825,1.8648,8347.55,107.75,1608.225,746.9005,1628.8958,3128.4225,1.5065,12.27,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.275,0,0.55,0.125,0,0.05,8.45,137,124.875,1.8673,8398.85,108.75,1637.575,746.797,1633.1663,3142.1675,1.4815,9.15,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.4,0,0.4,0.15,0,0.05,8.25,137.75,124.75,1.861,8226.6,106.1,1574.5,744.3675,1621.0475,3101.675,1.486,17.18,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.35,0,0.45,0.15,0,0.05,8.3,137.5,124.8,1.8635,8277.9,107.1,1603.85,744.264,1625.318,3115.42,1.461,16.72,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0,0.5,0.15,0,0.05,8.35,137.25,124.85,1.866,8329.2,108.1,1633.2,744.1605,1629.5885,3129.165,1.436,15.76,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.25,0,0.55,0.15,0,0.05,8.4,137,124.9,1.8685,8380.5,109.1,1662.55,744.057,1633.859,3142.91,1.411,14.42,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+Fe-Ni-Co-Cr,0.618,0.091,0.291,0,0,0,8.473,138.09,124.382,1.8518,8266.66,86.821,1401.594,759.5626,1605.1289,3077.2175,1.9271,10.53,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.527,0.091,0.382,0,0,0,8.564,137.635,124.473,1.8564,8360.026,88.641,1455.011,759.3742,1612.9012,3102.2334,1.8816,10.51,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.473,0.091,0.436,0,0,0,8.618,137.365,124.527,1.8591,8415.43,89.721,1486.709,759.2625,1617.5133,3117.078,1.8546,11.05,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.4,0.091,0.509,0,0,0,8.691,137,124.6,1.8627,8490.328,91.181,1529.56,759.1113,1623.7482,3137.1457,1.8181,11.06,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.364,0.091,0.545,0,0,0,8.727,136.82,124.636,1.8645,8527.264,91.901,1550.692,759.0368,1626.823,3147.0421,1.8001,10.92,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.327,0.091,0.582,0,0,0,8.764,136.635,124.673,1.8664,8565.226,92.641,1572.411,758.9602,1629.9832,3157.2134,1.7816,9.91,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.273,0.091,0.636,0,0,0,8.818,136.365,124.727,1.8691,8620.63,93.721,1604.109,758.8485,1634.5953,3172.058,1.7546,11.78,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.227,0.091,0.682,0,0,0,8.864,136.135,124.773,1.8714,8667.826,94.641,1631.111,758.7532,1638.5242,3184.7034,1.7316,12.55,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.582,0.091,0.291,0.036,0,0,8.401,138.09,124.418,1.8536,8240.236,87.325,1437.558,755.617,1606.1264,3078.2867,1.8256,11.07,1,10_50,,,,0.6182,0.0909,0.2909,0,0.618,0.091,0.291,0,,,,,,,,,,,
+,0.546,0.091,0.327,0.036,0,0,8.437,137.91,124.454,1.8554,8277.172,88.045,1458.69,755.5425,1609.2012,3088.1831,1.8076,10.71,1,10_50,,,,0.5273,0.0909,0.3818,0,0.527,0.091,0.382,0,,,,,,,,,,,
+,0.518,0.091,0.355,0.036,0,0,8.465,137.77,124.482,1.8568,8305.9,88.605,1475.126,755.4845,1611.5927,3095.8803,1.7936,10.42,1,10_50,,,,0.4727,0.0909,0.4364,0,0.473,0.091,0.436,0,,,,,,,,,,,
+,0.491,0.091,0.382,0.036,0,0,8.492,137.635,124.509,1.8582,8333.602,89.145,1490.975,755.4286,1613.8987,3103.3026,1.7801,10.75,1,10_50,,,,0.4,0.0909,0.5091,0,0.4,0.091,0.509,0,,,,,,,,,,,
+,0.464,0.091,0.409,0.036,0,0,8.519,137.5,124.536,1.8595,8361.304,89.685,1506.824,755.3727,1616.2048,3110.7249,1.7666,10.83,1,10_50,,,,0.3636,0.0909,0.5455,0,0.364,0.091,0.545,0,,,,,,,,,,,
+,0.437,0.091,0.436,0.036,0,0,8.546,137.365,124.563,1.8609,8389.006,90.225,1522.673,755.3169,1618.5109,3118.1472,1.7531,10.76,1,10_50,,,,0.3273,0.0909,0.5818,0,0.327,0.091,0.582,0,,,,,,,,,,,
+,0.4,0.091,0.473,0.036,0,0,8.583,137.18,124.6,1.8627,8426.968,90.965,1544.392,755.2403,1621.671,3128.3185,1.7346,11.02,1,10_50,,,,0.2727,0.0909,0.6364,0,0.273,0.091,0.636,0,,,,,,,,,,,
+,0.364,0.091,0.509,0.036,0,0,8.619,137,124.636,1.8645,8463.904,91.685,1565.524,755.1657,1624.7458,3138.2149,1.7166,11.16,1,10_50,,,,0.2273,0.0909,0.6818,0,0.227,0.091,0.682,0,,,,,,,,,,,
+,0.328,0.091,0.545,0.036,0,0,8.655,136.82,124.672,1.8663,8500.84,92.405,1586.656,755.0912,1627.8206,3148.1113,1.6986,10.76,1,10_50,,,,0.5818,0.0909,0.2909,0.0364,0.582,0.091,0.291,0.036,,,,,,,,,,,
+,0.273,0.091,0.6,0.036,0,0,8.71,136.545,124.727,1.8691,8557.27,93.505,1618.941,754.9774,1632.5181,3163.2308,1.6711,11.58,1,10_50,,,,0.5455,0.0909,0.3273,0.0364,0.545,0.091,0.327,0.036,,,,,,,,,,,
+,0.228,0.091,0.645,0.036,0,0,8.755,136.32,124.772,1.8713,8603.44,94.405,1645.356,754.8842,1636.3616,3175.6013,1.6486,12.22,1,10_50,,,,0.5182,0.0909,0.3546,0.0364,0.518,0.091,0.355,0.036,,,,,,,,,,,
+,0.423,0.091,0.436,0.05,0,0,8.518,137.365,124.577,1.8616,8378.73,90.421,1536.659,753.7825,1618.8988,3118.563,1.7136,10.95,1,10_50,,,,0.4909,0.0909,0.3818,0.0364,0.491,0.091,0.382,0.036,,,,,,,,,,,
+,0.518,0.091,0.327,0.064,0,0,8.381,137.91,124.482,1.8568,8256.62,88.437,1486.662,752.4737,1609.9771,3089.0147,1.7286,10.89,1,10_50,,,,0.4636,0.0909,0.4091,0.0364,0.464,0.091,0.409,0.036,,,,,,,,,,,
+,0.49,0.091,0.355,0.064,0,0,8.409,137.77,124.51,1.8582,8285.348,88.997,1503.098,752.4157,1612.3685,3096.7119,1.7146,10.69,1,10_50,,,,0.4364,0.0909,0.4364,0.0364,0.436,0.091,0.436,0.036,,,,,,,,,,,
+,0.463,0.091,0.382,0.064,0,0,8.436,137.635,124.537,1.8596,8313.05,89.537,1518.947,752.3598,1614.6746,3104.1342,1.7011,10.89,1,10_50,,,,0.4,0.0909,0.4727,0.0364,0.4,0.091,0.473,0.036,,,,,,,,,,,
+,0.436,0.091,0.409,0.064,0,0,8.463,137.5,124.564,1.8609,8340.752,90.077,1534.796,752.3039,1616.9807,3111.5565,1.6876,11.07,1,10_50,,,,0.3636,0.0909,0.5091,0.0364,0.364,0.091,0.509,0.036,,,,,,,,,,,
+,0.409,0.091,0.436,0.064,0,0,8.49,137.365,124.591,1.8623,8368.454,90.617,1550.645,752.2481,1619.2868,3118.9788,1.6741,10.92,1,10_50,,,,0.3273,0.0909,0.5455,0.0364,0.327,0.091,0.545,0.036,,,,,,,,,,,
+,0.372,0.091,0.473,0.064,0,0,8.527,137.18,124.628,1.8641,8406.416,91.357,1572.364,752.1715,1622.4469,3129.1501,1.6556,8.16,1,10_50,,,,0.2727,0.0909,0.6,0.0364,0.273,0.091,0.6,0.036,,,,,,,,,,,
+,0.336,0.091,0.509,0.064,0,0,8.563,137,124.664,1.8659,8443.352,92.077,1593.496,752.0969,1625.5217,3139.0465,1.6376,9.24,1,10_50,,,,0.2273,0.0909,0.6455,0.0364,0.227,0.091,0.645,0.036,,,,,,,,,,,
+,0.272,0.091,0.573,0.064,0,0,8.627,136.68,124.728,1.8691,8509.016,93.357,1631.064,751.9645,1630.9879,3156.6401,1.6056,11.1,1,10_50,,,,0.4227,0.0909,0.4364,0.05,0.423,0.091,0.436,0.05,,,,,,,,,,,
+,0.227,0.091,0.618,0.064,0,0,8.672,136.455,124.773,1.8714,8555.186,94.257,1657.479,751.8713,1634.8314,3169.0106,1.5831,11.74,1,10_50,,,,0.5182,0.0909,0.3273,0.0636,0.518,0.091,0.327,0.064,,,,,,,,,,,
+,0.441,0.091,0.386,0.082,0,0,8.404,137.615,124.559,1.8607,8303.942,89.869,1539.277,750.3788,1615.515,3105.7684,1.6483,4.06,1,10_50,,,,0.4909,0.0909,0.3546,0.0636,0.491,0.091,0.355,0.064,,,,,,,,,,,
+(?**),0.422,0.091,0.405,0.082,0,0,8.423,137.52,124.578,1.8616,8323.436,90.249,1550.43,750.3394,1617.1378,3110.9915,1.6388,3.66,1,10_50,,,,0.4636,0.0909,0.3818,0.0636,0.464,0.091,0.382,0.064,,,,,,,,,,,
+,0.428,0.091,0.395,0.086,0,0,8.405,137.57,124.572,1.8613,8310.24,90.105,1548.556,749.9217,1616.3946,3108.3613,1.6326,2.2,1,10_50,,,,0.4364,0.0909,0.4091,0.0636,0.436,0.091,0.409,0.064,,,,,,,,,,,
+,0.423,0.091,0.4,0.086,0,0,8.41,137.545,124.577,1.8616,8315.37,90.205,1551.491,749.9114,1616.8216,3109.7358,1.6301,2.36,1,10_50,,,,0.4091,0.0909,0.4364,0.0636,0.409,0.091,0.436,0.064,,,,,,,,,,,
+,0.418,0.091,0.405,0.086,0,0,8.415,137.52,124.582,1.8618,8320.5,90.305,1554.426,749.901,1617.2487,3111.1103,1.6276,2.65,1,10_50,,,,0.3727,0.0909,0.4727,0.0636,0.373,0.091,0.473,0.064,,,,,,,,,,,
+,0.414,0.091,0.409,0.086,0,0,8.419,137.5,124.586,1.862,8324.604,90.385,1556.774,749.8927,1617.5903,3112.2099,1.6256,2.9,1,10_50,,,,0.3364,0.0909,0.5091,0.0636,0.336,0.091,0.509,0.064,,,,,,,,,,,
+,0.527,0.091,0.291,0.091,0,0,8.291,138.09,124.473,1.8564,8199.866,88.095,1492.503,749.589,1607.6505,3079.9202,1.6705,12.4,1,10_50,,,,0.2727,0.0909,0.5727,0.0636,0.273,0.091,0.573,0.064,,,,,,,,,,,
+,0.491,0.091,0.327,0.091,0,0,8.327,137.91,124.509,1.8582,8236.802,88.815,1513.635,749.5145,1610.7252,3089.8166,1.6525,12.05,1,10_50,,,,0.2273,0.0909,0.6182,0.0636,0.227,0.091,0.618,0.064,,,,,,,,,,,
+,0.463,0.091,0.355,0.091,0,0,8.355,137.77,124.537,1.8596,8265.53,89.375,1530.071,749.4565,1613.1167,3097.5138,1.6385,8.86,1,10_50,,,,0.4409,0.0909,0.3864,0.0818,0.441,0.091,0.386,0.082,,,,,,,,,,,
+,0.436,0.091,0.382,0.091,0,0,8.382,137.635,124.564,1.8609,8293.232,89.915,1545.92,749.4006,1615.4228,3104.9361,1.625,3.03,1,10_50,,,,0.4227,0.0909,0.4046,0.0818,0.423,0.091,0.405,0.082,,,,,,,,,,,
+,0.427,0.091,0.391,0.091,0,0,8.391,137.59,124.573,1.8614,8302.466,90.095,1551.203,749.382,1616.1915,3107.4102,1.6205,2.42,1,10_50,,,,0.4273,0.0909,0.3955,0.0864,0.427,0.091,0.395,0.086,,,,,,,,,,,
+,0.423,0.091,0.395,0.091,0,0,8.395,137.57,124.577,1.8616,8306.57,90.175,1553.551,749.3737,1616.5331,3108.5098,1.6185,2.47,1,10_50,,,,0.4227,0.0909,0.4,0.0864,0.423,0.091,0.4,0.086,,,,,,,,,,,
+,0.409,0.091,0.409,0.091,0,0,8.409,137.5,124.591,1.8623,8320.934,90.455,1561.769,749.3447,1617.7289,3112.3584,1.6115,3.24,1,10_50,,,,0.4182,0.0909,0.4046,0.0864,0.418,0.091,0.405,0.086,,,,,,,,,,,
+,0.395,0.091,0.423,0.091,0,0,8.423,137.43,124.605,1.863,8335.298,90.735,1569.987,749.3158,1618.9246,3116.207,1.6045,4.39,1,10_50,,,,0.4136,0.0909,0.4091,0.0864,0.414,0.091,0.409,0.086,,,,,,,,,,,
+,0.382,0.091,0.436,0.091,0,0,8.436,137.365,124.618,1.8636,8348.636,90.995,1577.618,749.2889,1620.0349,3119.7807,1.598,4.95,1,10_50,,,,0.5273,0.0909,0.2909,0.0909,0.527,0.091,0.291,0.091,,,,,,,,,,,
+,0.345,0.091,0.473,0.091,0,0,8.473,137.18,124.655,1.8655,8386.598,91.735,1599.337,749.2123,1623.1951,3129.952,1.5795,7.78,1,10_50,,,,0.4909,0.0909,0.3273,0.0909,0.491,0.091,0.327,0.091,,,,,,,,,,,
+,0.309,0.091,0.509,0.091,0,0,8.509,137,124.691,1.8673,8423.534,92.455,1620.469,749.1377,1626.2699,3139.8484,1.5615,8.78,1,10_50,,,,0.4636,0.0909,0.3546,0.0909,0.464,0.091,0.355,0.091,,,,,,,,,,,
+,0.273,0.091,0.545,0.091,0,0,8.545,136.82,124.727,1.8691,8460.47,93.175,1641.601,749.0632,1629.3446,3149.7448,1.5435,9.96,1,10_50,,,,0.4364,0.0909,0.3818,0.0909,0.436,0.091,0.382,0.091,,,,,,,,,,,
+,0.227,0.091,0.591,0.091,0,0,8.591,136.59,124.773,1.8714,8507.666,94.095,1668.603,748.968,1633.2735,3162.3902,1.5205,10.61,1,10_50,,,,0.4273,0.0909,0.3909,0.0909,0.427,0.091,0.391,0.091,,,,,,,,,,,
+,0.182,0.091,0.636,0.091,0,0,8.636,136.365,124.818,1.8736,8553.836,94.995,1695.018,748.8749,1637.1169,3174.7607,1.498,11.78,1,10_50,,,,0.4227,0.0909,0.3955,0.0909,0.423,0.091,0.395,0.091,,,,,,,,,,,
+,0.414,0.091,0.395,0.1,0,0,8.377,137.57,124.586,1.862,8299.964,90.301,1562.542,748.3873,1616.7825,3108.7771,1.5931,3.71,1,10_50,,,,0.4091,0.0909,0.4091,0.0909,0.409,0.091,0.409,0.091,,,,,,,,,,,
+,0.464,0.091,0.327,0.118,0,0,8.273,137.91,124.536,1.8595,8216.984,89.193,1540.608,746.5553,1611.4734,3090.6185,1.5763,17.01,1,10_50,,,,0.3955,0.0909,0.4227,0.0909,0.395,0.091,0.423,0.091,,,,,,,,,,,
+,0.436,0.091,0.355,0.118,0,0,8.301,137.77,124.564,1.8609,8245.712,89.753,1557.044,746.4973,1613.8649,3098.3157,1.5623,16.01,1,10_50,,,,0.3818,0.0909,0.4364,0.0909,0.382,0.091,0.436,0.091,,,,,,,,,,,
+,0.409,0.091,0.382,0.118,0,0,8.328,137.635,124.591,1.8623,8273.414,90.293,1572.893,746.4414,1616.171,3105.738,1.5488,10.51,1,10_50,,,,0.3455,0.0909,0.4727,0.0909,0.345,0.091,0.473,0.091,,,,,,,,,,,
+,0.382,0.091,0.409,0.118,0,0,8.355,137.5,124.618,1.8636,8301.116,90.833,1588.742,746.3855,1618.477,3113.1603,1.5353,8.08,1,10_50,,,,0.3091,0.0909,0.5091,0.0909,0.309,0.091,0.509,0.091,,,,,,,,,,,
+,0.355,0.091,0.436,0.118,0,0,8.382,137.365,124.645,1.865,8328.818,91.373,1604.591,746.3297,1620.7831,3120.5826,1.5218,6.37,1,10_50,,,,0.2727,0.0909,0.5455,0.0909,0.273,0.091,0.545,0.091,,,,,,,,,,,
+,0.336,0.091,0.455,0.118,0,0,8.401,137.27,124.664,1.8659,8348.312,91.753,1615.744,746.2903,1622.4059,3125.8057,1.5123,6.78,1,10_50,,,,0.2273,0.0909,0.5909,0.0909,0.227,0.091,0.591,0.091,,,,,,,,,,,
+,0.308,0.091,0.483,0.118,0,0,8.419,137.18,124.682,1.8668,8366.78,92.113,1626.31,746.2531,1623.9433,3130.7539,1.5033,7.27,1,10_50,,,,0.1818,0.0909,0.6364,0.0909,0.182,0.091,0.636,0.091,,,,,,,,,,,
+,0.282,0.091,0.509,0.118,0,0,8.455,137,124.718,1.8686,8403.716,92.833,1647.442,746.1785,1627.018,3140.6503,1.4853,8.36,1,10_50,,,,0.4136,0.0909,0.3955,0.1,0.414,0.091,0.395,0.1,,,,,,,,,,,
+,0.246,0.091,0.545,0.118,0,0,8.491,136.82,124.754,1.8704,8440.652,93.553,1668.574,746.104,1630.0928,3150.5467,1.4673,9.84,1,10_50,,,,0.4636,0.0909,0.3273,0.1182,0.464,0.091,0.327,0.118,,,,,,,,,,,
+,0.2,0.091,0.591,0.118,0,0,8.537,136.59,124.8,1.8727,8487.848,94.473,1695.576,746.0088,1634.0216,3163.1921,1.4443,14.39,1,10_50,,,,0.4364,0.0909,0.3546,0.1182,0.436,0.091,0.355,0.118,,,,,,,,,,,
+,0.346,0.091,0.436,0.127,0,0,8.364,137.365,124.654,1.8654,8322.212,91.499,1613.582,745.3433,1621.0325,3120.8499,1.4964,7.35,1,10_50,,,,0.4091,0.0909,0.3818,0.1182,0.409,0.091,0.382,0.118,,,,,,,,,,,
+,0.493,0.091,0.301,0.145,0,0,8.183,138.09,124.527,1.8591,8160.23,88.851,1546.449,743.6706,1609.1468,3081.524,1.5182,16.95,1,10_50,,,,0.3818,0.0909,0.4091,0.1182,0.382,0.091,0.409,0.118,,,,,,,,,,,
+,0.437,0.091,0.327,0.145,0,0,8.219,137.91,124.563,1.8609,8197.166,89.571,1567.581,743.5961,1612.2216,3091.4204,1.5002,17.73,1,10_50,,,,0.3546,0.0909,0.4364,0.1182,0.355,0.091,0.436,0.118,,,,,,,,,,,
+,0.409,0.091,0.355,0.145,0,0,8.247,137.77,124.591,1.8623,8225.894,90.131,1584.017,743.5381,1614.6131,3099.1176,1.4862,16.91,1,10_50,,,,0.3364,0.0909,0.4546,0.1182,0.336,0.091,0.455,0.118,,,,,,,,,,,
+,0.382,0.091,0.382,0.145,0,0,8.274,137.635,124.618,1.8636,8253.596,90.671,1599.866,743.4822,1616.9191,3106.5399,1.4727,15.64,1,10_50,,,,0.3182,0.0909,0.4727,0.1182,0.318,0.091,0.473,0.118,,,,,,,,,,,
+,0.328,0.091,0.436,0.145,0,0,8.328,137.365,124.672,1.8663,8309,91.751,1631.564,743.3705,1621.5313,3121.3845,1.4457,13.64,1,10_50,,,,0.2818,0.0909,0.5091,0.1182,0.282,0.091,0.509,0.118,,,,,,,,,,,
+,0.291,0.091,0.473,0.145,0,0,8.365,137.18,124.709,1.8682,8346.962,92.491,1653.283,743.2939,1624.6914,3131.5558,1.4272,9.07,1,10_50,,,,0.2455,0.0909,0.5455,0.1182,0.245,0.091,0.545,0.118,,,,,,,,,,,
+,0.273,0.091,0.491,0.145,0,0,8.383,137.09,124.727,1.8691,8365.43,92.851,1663.849,743.2566,1626.2288,3136.504,1.4182,9.46,1,10_50,,,,0.2,0.0909,0.5909,0.1182,0.2,0.091,0.591,0.118,,,,,,,,,,,
+,0.228,0.091,0.536,0.145,0,0,8.428,136.865,124.772,1.8713,8411.6,93.751,1690.264,743.1635,1630.0723,3148.8745,1.3957,10.37,1,10_50,,,,0.3455,0.0909,0.4364,0.1273,0.345,0.091,0.436,0.127,,,,,,,,,,,
+,0.155,0.091,0.609,0.145,0,0,8.501,136.5,124.845,1.875,8486.498,95.211,1733.115,743.0123,1636.3072,3168.9422,1.3592,14.63,1,10_50,,,,0.4727,0.0909,0.2909,0.1455,0.473,0.091,0.291,0.145,,,,,,,,,,,
+,0.227,0.091,0.527,0.155,0,0,8.399,136.91,124.773,1.8714,8395.026,93.711,1694.971,742.0861,1629.5807,3146.6974,1.372,10.9,1,10_50,,,,0.4364,0.0909,0.3273,0.1455,0.436,0.091,0.327,0.145,,,,,,,,,,,
+,0.272,0.091,0.473,0.164,0,0,8.327,137.18,124.728,1.8691,8333.016,92.757,1672.264,741.2115,1625.2179,3132.1201,1.3736,13.52,1,10_50,,,,0.4091,0.0909,0.3546,0.1455,0.409,0.091,0.355,0.145,,,,,,,,,,,
+,0.436,0.091,0.291,0.182,0,0,8.109,138.09,124.564,1.8609,8133.072,89.369,1583.412,739.6154,1610.1721,3082.6229,1.4138,17.25,1,10_50,,,,0.3818,0.0909,0.3818,0.1455,0.382,0.091,0.382,0.145,,,,,,,,,,,
+,0.4,0.091,0.327,0.182,0,0,8.145,137.91,124.6,1.8627,8170.008,90.089,1604.544,739.5409,1613.2468,3092.5193,1.3958,16.89,1,10_50,,,,0.3273,0.0909,0.4364,0.1455,0.327,0.091,0.436,0.145,,,,,,,,,,,
+,0.345,0.091,0.382,0.182,0,0,8.2,137.635,124.655,1.8655,8226.438,91.189,1636.829,739.427,1617.9444,3107.6388,1.3683,16.54,1,10_50,,,,0.2909,0.0909,0.4727,0.1455,0.291,0.091,0.473,0.145,,,,,,,,,,,
+,0.291,0.091,0.436,0.182,0,0,8.254,137.365,124.709,1.8682,8281.842,92.269,1668.527,739.3153,1622.5565,3122.4834,1.3413,16.01,1,10_50,,,,0.2727,0.0909,0.4909,0.1455,0.273,0.091,0.491,0.145,,,,,,,,,,,
+,0.254,0.091,0.473,0.182,0,0,8.291,137.18,124.746,1.87,8319.804,93.009,1690.246,739.2387,1625.7167,3132.6547,1.3228,13.81,1,10_50,,,,0.2273,0.0909,0.5364,0.1455,0.227,0.091,0.536,0.145,,,,,,,,,,,
+,0.204,0.091,0.523,0.182,0,0,8.341,136.93,124.796,1.8725,8371.104,94.009,1719.596,739.1352,1629.9872,3146.3997,1.2978,18.06,1,10_50,,,,0.1546,0.0909,0.6091,0.1455,0.155,0.091,0.609,0.145,,,,,,,,,,,
+,0.136,0.091,0.591,0.182,0,0,8.409,136.59,124.864,1.8759,8440.872,95.369,1759.512,738.9944,1635.7951,3165.0929,1.2638,17.84,1,10_50,,,,0.2273,0.0909,0.5273,0.1546,0.227,0.091,0.527,0.155,,,,,,,,,,,
+,0.666,0.167,0.167,0,0,0,8.501,138.33,124.334,1.8517,8218.02,85.177,1370.378,757.8942,1608.9818,3076.3875,1.866,10.46,1,10_50,,,,0.2727,0.0909,0.4727,0.1636,0.273,0.091,0.473,0.164,,,,,,,,,,,
+,0.583,0.167,0.25,0,0,0,8.584,137.915,124.417,1.8559,8303.178,86.837,1419.099,757.7224,1616.0709,3099.2042,1.8245,10.87,1,10_50,,,,0.4364,0.0909,0.2909,0.1818,0.436,0.091,0.291,0.182,,,,,,,,,,,
+,0.475,0.167,0.358,0,0,0,8.692,137.375,124.525,1.8613,8413.986,88.997,1482.495,757.4988,1625.2951,3128.8934,1.7705,11.08,1,10_50,,,,0.4,0.0909,0.3273,0.1818,0.4,0.091,0.327,0.182,,,,,,,,,,,
+,0.375,0.167,0.458,0,0,0,8.792,136.875,124.625,1.8663,8516.586,90.997,1541.195,757.2918,1633.8361,3156.3834,1.7205,11.52,1,10_50,,,,0.3455,0.0909,0.3818,0.1818,0.345,0.091,0.382,0.182,,,,,,,,,,,
+,0.333,0.167,0.5,0,0,0,8.834,136.665,124.667,1.8684,8559.678,91.837,1565.849,757.2049,1637.4234,3167.9292,1.6995,12.05,1,10_50,,,,0.2909,0.0909,0.4364,0.1818,0.291,0.091,0.436,0.182,,,,,,,,,,,
+,0.291,0.167,0.542,0,0,0,8.876,136.455,124.709,1.8705,8602.77,92.677,1590.503,757.118,1641.0106,3179.475,1.6785,12.37,1,10_50,,,,0.2546,0.0909,0.4727,0.1818,0.255,0.091,0.473,0.182,,,,,,,,,,,
+,0.567,0.167,0.233,0.033,0,0,8.501,138,124.433,1.8567,8261.514,86.959,1442.087,754.1408,1615.5333,3095.511,1.7399,11.66,1,10_50,,,,0.2046,0.0909,0.5227,0.1818,0.205,0.091,0.523,0.182,,,,,,,,,,,
+,0.521,0.167,0.279,0.033,0,0,8.547,137.77,124.479,1.859,8308.71,87.879,1469.089,754.0456,1619.4622,3108.1564,1.7169,11.35,1,10_50,,,,0.1364,0.0909,0.5909,0.1818,0.136,0.091,0.591,0.182,,,,,,,,,,,
+,0.5,0.167,0.3,0.033,0,0,8.568,137.665,124.5,1.86,8330.256,88.299,1481.416,754.0021,1621.2558,3113.9293,1.7064,10.65,1,10_50,,,,0.6667,0.1667,0.1667,0,0.667,0.167,0.167,0,,,,,,,,,,,
+,0.458,0.167,0.342,0.033,0,0,8.61,137.455,124.542,1.8621,8373.348,89.139,1506.07,753.9152,1624.843,3125.4751,1.6854,9.22,1,10_50,,,,0.5833,0.1667,0.25,0,0.583,0.167,0.25,0,,,,,,,,,,,
+,0.437,0.167,0.363,0.033,0,0,8.631,137.35,124.563,1.8632,8394.894,89.559,1518.397,753.8717,1626.6366,3131.248,1.6749,8.61,1,10_50,,,,0.475,0.1667,0.3583,0,0.475,0.167,0.358,0,,,,,,,,,,,
+,0.417,0.167,0.383,0.033,0,0,8.651,137.25,124.583,1.8642,8415.414,89.959,1530.137,753.8303,1628.3448,3136.746,1.6649,8.21,1,10_50,,,,0.375,0.1667,0.4583,0,0.375,0.167,0.458,0,,,,,,,,,,,
+,0.396,0.167,0.404,0.033,0,0,8.672,137.145,124.604,1.8652,8436.96,90.379,1542.464,753.7868,1630.1384,3142.5189,1.6544,8.87,1,10_50,,,,0.3333,0.1667,0.5,0,0.333,0.167,0.5,0,,,,,,,,,,,
+,0.358,0.167,0.442,0.033,0,0,8.71,136.955,124.642,1.8671,8475.948,91.139,1564.77,753.7082,1633.384,3152.9651,1.6354,11.21,1,10_50,,,,0.2917,0.1667,0.5417,0,0.292,0.167,0.542,0,,,,,,,,,,,
+,0.317,0.167,0.483,0.033,0,0,8.751,136.75,124.683,1.8692,8518.014,91.959,1588.837,753.6233,1636.8858,3164.236,1.6149,11.95,1,10_50,,,,0.5667,0.1667,0.2333,0.0333,0.567,0.167,0.233,0.033,,,,,,,,,,,
+,0.296,0.167,0.504,0.033,0,0,8.772,136.645,124.704,1.8702,8539.56,92.379,1601.164,753.5798,1638.6794,3170.0089,1.6044,11.75,1,10_50,,,,0.5208,0.1667,0.2792,0.0333,0.521,0.167,0.279,0.033,,,,,,,,,,,
+,0.575,0.167,0.2,0.058,0,0,8.418,138.165,124.425,1.8563,8209.306,86.649,1447.691,751.4691,1613.4075,3087.1818,1.6859,11.02,1,10_50,,,,0.5,0.1667,0.3,0.0333,0.5,0.167,0.3,0.033,,,,,,,,,,,
+,0.529,0.167,0.246,0.058,0,0,8.464,137.935,124.471,1.8586,8256.502,87.569,1474.693,751.3739,1617.3364,3099.8272,1.6629,9.25,1,10_50,,,,0.4583,0.1667,0.3417,0.0333,0.458,0.167,0.342,0.033,,,,,,,,,,,
+,0.508,0.167,0.267,0.058,0,0,8.485,137.83,124.492,1.8596,8278.048,87.989,1487.02,751.3304,1619.13,3105.6001,1.6524,7.75,1,10_50,,,,0.4375,0.1667,0.3625,0.0333,0.438,0.167,0.363,0.033,,,,,,,,,,,
+,0.487,0.167,0.288,0.058,0,0,8.506,137.725,124.513,1.8607,8299.594,88.409,1499.347,751.2869,1620.9236,3111.373,1.6419,3.23,1,10_50,,,,0.4167,0.1667,0.3833,0.0333,0.417,0.167,0.383,0.033,,,,,,,,,,,
+,0.467,0.167,0.308,0.058,0,0,8.526,137.625,124.533,1.8617,8320.114,88.809,1511.087,751.2455,1622.6318,3116.871,1.6319,5.79,1,10_50,,,,0.3958,0.1667,0.4042,0.0333,0.396,0.167,0.404,0.033,,,,,,,,,,,
+,0.426,0.187,0.329,0.058,0,0,8.547,137.52,124.554,1.8627,8341.66,89.229,1523.414,751.2021,1624.4254,3122.6439,1.6214,6.82,1,10_50,,,,0.3583,0.1667,0.4417,0.0333,0.358,0.167,0.442,0.033,,,,,,,,,,,
+,0.404,0.167,0.371,0.058,0,0,8.589,137.31,124.596,1.8648,8384.752,90.069,1548.068,751.1151,1628.0126,3134.1897,1.6004,9.48,1,10_50,,,,0.3167,0.1667,0.4833,0.0333,0.317,0.167,0.483,0.033,,,,,,,,,,,
+,0.383,0.167,0.392,0.058,0,0,8.61,137.205,124.617,1.8659,8406.298,90.489,1560.395,751.0717,1629.8063,3139.9626,1.5899,9.18,1,10_50,,,,0.2958,0.1667,0.5042,0.0333,0.296,0.167,0.504,0.033,,,,,,,,,,,
+,0.362,0.167,0.413,0.058,0,0,8.631,137.1,124.638,1.8669,8427.844,90.909,1572.722,751.0282,1631.5999,3145.7355,1.5794,9.78,1,10_50,,,,0.575,0.1667,0.2,0.0583,0.575,0.167,0.2,0.058,,,,,,,,,,,
+,0.333,0.167,0.442,0.058,0,0,8.66,136.955,124.667,1.8684,8457.598,91.489,1589.745,750.9682,1634.0768,3153.7076,1.5649,10.77,1,10_50,,,,0.5292,0.1667,0.2458,0.0583,0.529,0.167,0.246,0.058,,,,,,,,,,,
+,0.283,0.167,0.492,0.058,0,0,8.71,136.705,124.717,1.8709,8508.898,92.489,1619.095,750.8647,1638.3473,3167.4526,1.5399,11.03,1,10_50,,,,0.5083,0.1667,0.2667,0.0583,0.508,0.167,0.267,0.058,,,,,,,,,,,
+,0.512,0.167,0.25,0.071,0,0,8.442,137.915,124.488,1.8594,8251.064,87.831,1490.028,749.9408,1618.0383,3101.3129,1.6242,2.04,1,10_50,,,,0.4875,0.1667,0.2875,0.0583,0.488,0.167,0.288,0.058,,,,,,,,,,,
+,0.501,0.167,0.261,0.071,0,0,8.453,137.86,124.499,1.86,8262.35,88.051,1496.485,749.918,1618.9778,3104.3368,1.6187,1.69,1,10_50,,,,0.4667,0.1667,0.3083,0.0583,0.467,0.167,0.308,0.058,,,,,,,,,,,
+,0.5,0.167,0.258,0.075,0,0,8.442,137.875,124.5,1.86,8256.336,88.047,1498.72,749.4858,1618.8324,3103.6309,1.609,2.88,1,10_50,,,,0.4458,0.1667,0.3292,0.0583,0.446,0.167,0.329,0.058,,,,,,,,,,,
+,0.583,0.167,0.167,0.083,0,0,8.335,138.33,124.417,1.8559,8157.098,86.339,1453.295,748.7974,1611.2818,3078.8526,1.6319,17.91,1,10_50,,,,0.4042,0.1667,0.3708,0.0583,0.404,0.167,0.371,0.058,,,,,,,,,,,
+,0.562,0.167,0.198,0.073,0,0,8.356,138.225,124.438,1.8569,8178.644,86.759,1465.622,748.7539,1613.0754,3084.6255,1.6214,16.18,1,10_50,,,,0.3833,0.1667,0.3917,0.0583,0.383,0.167,0.392,0.058,,,,,,,,,,,
+,0.542,0.167,0.208,0.083,0,0,8.376,138.125,124.458,1.8579,8199.164,87.159,1477.362,748.7125,1614.7836,3090.1235,1.6114,12.01,1,10_50,,,,0.3625,0.1667,0.4125,0.0583,0.363,0.167,0.413,0.058,,,,,,,,,,,
+,0.517,0.167,0.233,0.083,0,0,8.401,138,124.483,1.8592,8224.814,87.659,1492.037,748.6608,1616.9188,3096.996,1.5989,5.57,1,10_50,,,,0.3333,0.1667,0.4417,0.0583,0.333,0.167,0.442,0.058,,,,,,,,,,,
+,0.496,0.167,0.254,0.083,0,0,8.422,137.895,124.504,1.8602,8246.36,88.079,1504.364,748.6173,1618.7124,3102.7689,1.5884,3.73,1,10_50,,,,0.2833,0.1667,0.4917,0.0583,0.283,0.167,0.492,0.058,,,,,,,,,,,
+,0.475,0.167,0.275,0.083,0,0,8.443,137.79,124.525,1.8613,8267.906,88.499,1516.691,748.5738,1620.506,3108.5418,1.5779,3.88,1,10_50,,,,0.5125,0.1667,0.25,0.0708,0.513,0.167,0.25,0.071,,,,,,,,,,,
+,0.454,0.167,0.296,0.083,0,0,8.464,137.685,124.546,1.8623,8289.452,88.919,1529.018,748.5304,1622.2996,3114.3147,1.5674,4.91,1,10_50,,,,0.5017,0.1667,0.2608,0.0708,0.502,0.167,0.261,0.071,,,,,,,,,,,
+,0.433,0.167,0.317,0.083,0,0,8.485,137.58,124.567,1.8634,8310.998,89.339,1541.345,748.4869,1624.0933,3120.0876,1.5569,5.98,1,10_50,,,,0.5,0.1667,0.2583,0.075,0.5,0.167,0.258,0.075,,,,,,,,,,,
+,0.412,0.167,0.338,0.083,0,0,8.506,137.475,124.588,1.8644,8332.544,89.759,1553.672,748.4434,1625.8869,3125.8605,1.5464,6.41,1,10_50,,,,0.5833,0.1667,0.1667,0.0833,0.583,0.167,0.167,0.083,,,,,,,,,,,
+,0.392,0.167,0.358,0.083,0,0,8.526,137.375,124.608,1.8654,8353.064,90.159,1565.412,748.402,1627.5951,3131.3585,1.5364,7.81,1,10_50,,,,0.5625,0.1667,0.1875,0.0833,0.563,0.167,0.188,0.083,,,,,,,,,,,
+,0.371,0.167,0.379,0.083,0,0,8.547,137.27,124.629,1.8665,8374.61,90.579,1577.739,748.3586,1629.3887,3137.1314,1.5259,7.77,1,10_50,,,,0.5417,0.1667,0.2083,0.0833,0.542,0.167,0.208,0.083,,,,,,,,,,,
+,0.333,0.167,0.417,0.083,0,0,8.585,137.08,124.667,1.8684,8413.598,91.339,1600.045,748.2799,1632.6343,3147.5776,1.5069,9.41,1,10_50,,,,0.5167,0.1667,0.2333,0.0833,0.517,0.167,0.233,0.083,,,,,,,,,,,
+,0.292,0.167,0.458,0.083,0,0,8.626,136.875,124.708,1.8704,8455.664,92.159,1624.112,748.195,1636.1361,3158.8485,1.4864,10.57,1,10_50,,,,0.4958,0.1667,0.2542,0.0833,0.496,0.167,0.254,0.083,,,,,,,,,,,
+,0.25,0.167,0.5,0.083,0,0,8.668,136.665,124.75,1.8725,8498.756,92.999,1648.766,748.1081,1639.7233,3170.3943,1.4654,12,1,10_50,,,,0.475,0.1667,0.275,0.0833,0.475,0.167,0.275,0.083,,,,,,,,,,,
+,0.558,0.167,0.167,0.108,0,0,8.285,138.33,124.442,1.8571,8138.748,86.689,1478.27,746.0574,1611.9745,3079.5951,1.5614,17.62,1,10_50,,,,0.4542,0.1667,0.2958,0.0833,0.454,0.167,0.296,0.083,,,,,,,,,,,
+,0.529,0.167,0.196,0.108,0,0,8.314,138.185,124.471,1.8586,8168.502,87.269,1495.293,745.9974,1614.4514,3087.5672,1.5469,16.23,1,10_50,,,,0.4333,0.1667,0.3167,0.0833,0.433,0.167,0.317,0.083,,,,,,,,,,,
+,0.504,0.167,0.221,0.108,0,0,8.339,138.06,124.496,1.8598,8194.152,87.769,1509.968,745.9456,1616.5866,3094.4397,1.5344,13.4,1,10_50,,,,0.4125,0.1667,0.3375,0.0833,0.413,0.167,0.338,0.083,,,,,,,,,,,
+,0.473,0.167,0.252,0.108,0,0,8.36,137.955,124.517,1.8609,8215.698,88.189,1522.295,745.9022,1618.3803,3100.2126,1.5239,10.47,1,10_50,,,,0.3917,0.1667,0.3583,0.0833,0.392,0.167,0.358,0.083,,,,,,,,,,,
+,0.462,0.167,0.263,0.108,0,0,8.381,137.85,124.538,1.8619,8237.244,88.609,1534.622,745.8587,1620.1739,3105.9855,1.5134,7.66,1,10_50,,,,0.3708,0.1667,0.3792,0.0833,0.371,0.167,0.379,0.083,,,,,,,,,,,
+,0.422,0.167,0.303,0.108,0,0,8.401,137.75,124.558,1.8629,8257.764,89.009,1546.362,745.8173,1621.8821,3111.4835,1.5034,6.88,1,10_50,,,,0.3333,0.1667,0.4167,0.0833,0.333,0.167,0.417,0.083,,,,,,,,,,,
+,0.421,0.167,0.304,0.108,0,0,8.422,137.645,124.579,1.864,8279.31,89.429,1558.689,745.7738,1623.6757,3117.2564,1.4929,5.95,1,10_50,,,,0.2917,0.1667,0.4583,0.0833,0.292,0.167,0.458,0.083,,,,,,,,,,,
+,0.4,0.167,0.325,0.108,0,0,8.443,137.54,124.6,1.865,8300.856,89.849,1571.016,745.7303,1625.4693,3123.0293,1.4824,6.17,1,10_50,,,,0.25,0.1667,0.5,0.0833,0.25,0.167,0.5,0.083,,,,,,,,,,,
+,0.379,0.167,0.346,0.108,0,0,8.464,137.435,124.621,1.8661,8322.402,90.269,1583.343,745.6869,1627.2629,3128.8022,1.4719,7.72,1,10_50,,,,0.5583,0.1667,0.1667,0.1083,0.558,0.167,0.167,0.108,,,,,,,,,,,
+,0.358,0.167,0.367,0.108,0,0,8.485,137.33,124.642,1.8671,8343.948,90.689,1595.67,745.6434,1629.0565,3134.5751,1.4614,7.79,1,10_50,,,,0.5292,0.1667,0.1958,0.1083,0.529,0.167,0.196,0.108,,,,,,,,,,,
+,0.337,0.167,0.388,0.108,0,0,8.506,137.225,124.663,1.8682,8365.494,91.109,1607.997,745.5999,1630.8501,3140.348,1.4509,8.34,1,10_50,,,,0.5042,0.1667,0.2208,0.1083,0.504,0.167,0.221,0.108,,,,,,,,,,,
+,0.292,0.167,0.433,0.108,0,0,8.551,137,124.708,1.8704,8411.664,92.009,1634.412,745.5068,1634.6936,3152.7185,1.4284,9.64,1,10_50,,,,0.4833,0.1667,0.2417,0.1083,0.483,0.167,0.242,0.108,,,,,,,,,,,
+,0.258,0.167,0.467,0.108,0,0,8.585,136.83,124.742,1.8721,8446.548,92.689,1654.37,745.4364,1637.5975,3162.0651,1.4114,10.8,1,10_50,,,,0.4625,0.1667,0.2625,0.1083,0.463,0.167,0.263,0.108,,,,,,,,,,,
+,0.492,0.167,0.208,0.133,0,0,8.276,138.125,124.508,1.8604,8162.464,87.859,1527.312,743.2325,1616.1691,3091.6085,1.4704,16.4,1,10_50,,,,0.4417,0.1667,0.2833,0.1083,0.442,0.167,0.283,0.108,,,,,,,,,,,
+,0.471,0.167,0.229,0.133,0,0,8.297,138.02,124.529,1.8615,8184.01,88.279,1539.639,743.1891,1617.9627,3097.3814,1.4599,16.2,1,10_50,,,,0.4208,0.1667,0.3042,0.1083,0.421,0.167,0.304,0.108,,,,,,,,,,,
+,0.45,0.167,0.25,0.133,0,0,8.318,137.915,124.55,1.8625,8205.556,88.699,1551.966,743.1456,1619.7563,3103.1543,1.4494,14.68,1,10_50,,,,0.4,0.1667,0.325,0.1083,0.4,0.167,0.325,0.108,,,,,,,,,,,
+,0.429,0.167,0.271,0.133,0,0,8.339,137.81,124.571,1.8636,8227.102,89.119,1564.293,743.1021,1621.5499,3108.9272,1.4389,13.41,1,10_50,,,,0.3792,0.1667,0.3458,0.1083,0.379,0.167,0.346,0.108,,,,,,,,,,,
+,0.408,0.167,0.292,0.133,0,0,8.36,137.705,124.592,1.8646,8248.648,89.539,1576.62,743.0587,1623.3435,3114.7001,1.4284,9.64,1,10_50,,,,0.3583,0.1667,0.3667,0.1083,0.358,0.167,0.367,0.108,,,,,,,,,,,
+,0.387,0.167,0.313,0.133,0,0,8.381,137.6,124.613,1.8657,8270.194,89.959,1588.947,743.0152,1625.1371,3120.473,1.4179,9.61,1,10_50,,,,0.3375,0.1667,0.3875,0.1083,0.338,0.167,0.388,0.108,,,,,,,,,,,
+,0.367,0.167,0.333,0.133,0,0,8.401,137.5,124.633,1.8667,8290.714,90.359,1600.687,742.9738,1626.8453,3125.971,1.4079,9.69,1,10_50,,,,0.2917,0.1667,0.4333,0.1083,0.292,0.167,0.433,0.108,,,,,,,,,,,
+,0.346,0.167,0.354,0.133,0,0,8.422,137.395,124.654,1.8677,8312.26,90.779,1613.014,742.9303,1628.6389,3131.7439,1.3974,8.92,1,10_50,,,,0.2583,0.1667,0.4667,0.1083,0.258,0.167,0.467,0.108,,,,,,,,,,,
+,0.325,0.167,0.375,0.133,0,0,8.443,137.29,124.675,1.8688,8333.806,91.199,1625.341,742.8868,1630.4325,3137.5168,1.3869,8.93,1,10_50,,,,0.4917,0.1667,0.2083,0.1333,0.492,0.167,0.208,0.133,,,,,,,,,,,
+,0.308,0.167,0.392,0.133,0,0,8.46,137.205,124.692,1.8696,8351.248,91.539,1635.32,742.8517,1631.8845,3142.1901,1.3784,9.09,1,10_50,,,,0.4708,0.1667,0.2292,0.1333,0.471,0.167,0.229,0.133,,,,,,,,,,,
+,0.292,0.167,0.408,0.133,0,0,8.476,137.125,124.708,1.8704,8367.664,91.859,1644.712,742.8185,1633.2511,3146.5885,1.3704,9.68,1,10_50,,,,0.45,0.1667,0.25,0.1333,0.45,0.167,0.25,0.133,,,,,,,,,,,
+,0.246,0.167,0.454,0.133,0,0,8.522,136.895,124.754,1.8727,8414.86,92.779,1671.714,742.7233,1637.1799,3159.2339,1.3474,10.54,1,10_50,,,,0.4292,0.1667,0.2708,0.1333,0.429,0.167,0.271,0.133,,,,,,,,,,,
+,0.508,0.167,0.167,0.158,0,0,8.185,138.33,124.492,1.8596,8102.048,87.389,1528.22,740.5774,1613.36,3081.0801,1.4204,16.84,1,10_50,,,,0.4083,0.1667,0.2917,0.1333,0.408,0.167,0.292,0.133,,,,,,,,,,,
+,0.458,0.167,0.217,0.158,0,0,8.235,138.08,124.542,1.8621,8153.348,88.389,1557.57,740.4739,1617.6305,3094.8251,1.3954,15.98,1,10_50,,,,0.3875,0.1667,0.3125,0.1333,0.388,0.167,0.313,0.133,,,,,,,,,,,
+,0.487,0.167,0.178,0.158,0,0,8.256,137.975,124.563,1.8632,8174.894,88.809,1569.897,740.4304,1619.4241,3100.598,1.3849,15.7,1,10_50,,,,0.3667,0.1667,0.3333,0.1333,0.367,0.167,0.333,0.133,,,,,,,,,,,
+,0.417,0.167,0.258,0.158,0,0,8.276,137.875,124.583,1.8642,8195.414,89.209,1581.637,740.389,1621.1323,3106.096,1.3749,15.31,1,10_50,,,,0.3458,0.1667,0.3542,0.1333,0.346,0.167,0.354,0.133,,,,,,,,,,,
+,0.375,0.167,0.3,0.158,0,0,8.318,137.665,124.625,1.8663,8238.506,90.049,1606.291,740.3021,1624.7195,3117.6418,1.3539,14.61,1,10_50,,,,0.325,0.1667,0.375,0.1333,0.325,0.167,0.375,0.133,,,,,,,,,,,
+,0.333,0.167,0.342,0.158,0,0,8.36,137.455,124.667,1.8684,8281.598,90.889,1630.945,740.2152,1628.3068,3129.1876,1.3329,14.09,1,10_50,,,,0.3083,0.1667,0.3917,0.1333,0.308,0.167,0.392,0.133,,,,,,,,,,,
+,0.312,0.167,0.363,0.158,0,0,8.381,137.35,124.688,1.8694,8303.144,91.309,1643.272,740.1717,1630.1004,3134.9605,1.3224,12.22,1,10_50,,,,0.2917,0.1667,0.4083,0.1333,0.292,0.167,0.408,0.133,,,,,,,,,,,
+,0.292,0.167,0.383,0.158,0,0,8.401,137.25,124.708,1.8704,8323.664,91.709,1655.012,740.1303,1631.8086,3140.4585,1.3124,12.33,1,10_50,,,,0.2458,0.1667,0.4542,0.1333,0.246,0.167,0.454,0.133,,,,,,,,,,,
+,0.254,0.167,0.421,0.158,0,0,8.439,137.06,124.746,1.8723,8362.652,92.469,1677.318,740.0516,1635.0541,3150.9047,1.2934,10.86,1,10_50,,,,0.5083,0.1667,0.1667,0.1583,0.508,0.167,0.167,0.158,,,,,,,,,,,
+,0.212,0.167,0.463,0.158,0,0,8.481,136.85,124.788,1.8744,8405.744,93.309,1701.972,739.9647,1638.6414,3162.4505,1.2724,12.39,1,10_50,,,,0.4583,0.1667,0.2167,0.1583,0.458,0.167,0.217,0.158,,,,,,,,,,,
+,0.692,0.231,0.077,0,0,0,8.539,138.46,124.308,1.8523,8191.856,84.081,1352.556,756.4594,1613.4581,3079.6529,1.8073,10.58,1,10_50,,,,0.4375,0.1667,0.2375,0.1583,0.438,0.167,0.238,0.158,,,,,,,,,,,
+,0.615,0.231,0.154,0,0,0,8.616,138.075,124.385,1.8562,8270.858,85.621,1397.755,756.3,1620.0347,3100.8202,1.7688,10.93,1,10_50,,,,0.4167,0.1667,0.2583,0.1583,0.417,0.167,0.258,0.158,,,,,,,,,,,
+,0.584,0.231,0.185,0,0,0,8.647,137.92,124.416,1.8577,8302.664,86.241,1415.952,756.2358,1622.6824,3109.3421,1.7533,11.12,1,10_50,,,,0.375,0.1667,0.3,0.1583,0.375,0.167,0.3,0.158,,,,,,,,,,,
+,0.538,0.231,0.231,0,0,0,8.693,137.69,124.462,1.86,8349.86,87.161,1442.954,756.1406,1626.6113,3121.9875,1.7303,11.82,1,10_50,,,,0.3333,0.1667,0.3417,0.1583,0.333,0.167,0.342,0.158,,,,,,,,,,,
+,0.5,0.231,0.269,0,0,0,8.731,137.5,124.5,1.8619,8388.848,87.921,1465.26,756.0619,1629.8568,3132.4337,1.7113,11.92,1,10_50,,,,0.3125,0.1667,0.3625,0.1583,0.313,0.167,0.363,0.158,,,,,,,,,,,
+,0.461,0.231,0.308,0,0,0,8.77,137.305,124.539,1.8639,8428.862,88.701,1488.153,755.9812,1633.1878,3143.1548,1.6918,12.38,1,10_50,,,,0.2917,0.1667,0.3833,0.1583,0.292,0.167,0.383,0.158,,,,,,,,,,,
+,0.431,0.231,0.338,0,0,0,8.8,137.155,124.569,1.8654,8459.642,89.301,1505.763,755.9191,1635.7501,3151.4018,1.6768,10.54,1,10_50,,,,0.2542,0.1667,0.4208,0.1583,0.254,0.167,0.421,0.158,,,,,,,,,,,
+,0.364,0.231,0.405,0,0,0,8.847,136.92,124.616,1.8677,8507.864,90.241,1533.352,755.8218,1639.7644,3164.3221,1.6533,11.64,1,10_50,,,,0.2125,0.1667,0.4625,0.1583,0.213,0.167,0.463,0.158,,,,,,,,,,,
+,0.346,0.231,0.423,0,0,0,8.885,136.73,124.654,1.8696,8546.852,91.001,1555.658,755.7432,1643.01,3174.7683,1.6343,11.9,1,10_50,,,,0.6923,0.2308,0.0769,0,0.692,0.231,0.077,0,,,,,,,,,,,
+,0.631,0.231,0.115,0.023,0,0,8.531,138.27,124.369,1.8554,8213.962,85.163,1397.839,753.8599,1617.341,3090.7822,1.7234,10.83,1,10_50,,,,0.6154,0.2308,0.1539,0,0.615,0.231,0.154,0,,,,,,,,,,,
+,0.592,0.231,0.154,0.023,0,0,8.57,138.075,124.408,1.8573,8253.976,85.943,1420.732,753.7792,1620.672,3101.5033,1.7039,10.59,1,10_50,,,,0.5846,0.2308,0.1846,0,0.585,0.231,0.185,0,,,,,,,,,,,
+,0.554,0.231,0.192,0.023,0,0,8.608,137.885,124.446,1.8592,8292.964,86.703,1443.038,753.7005,1623.9176,3111.9495,1.6849,10.08,1,10_50,,,,0.5385,0.2308,0.2308,0,0.538,0.231,0.231,0,,,,,,,,,,,
+,0.5,0.231,0.246,0.023,0,0,8.662,137.615,124.5,1.8619,8348.368,87.783,1474.736,753.5888,1628.5297,3126.7941,1.6579,8.25,1,10_50,,,,0.5,0.2308,0.2692,0,0.5,0.231,0.269,0,,,,,,,,,,,
+,0.461,0.231,0.285,0.023,0,0,8.701,137.42,124.539,1.8639,8388.382,88.563,1497.629,753.508,1631.8607,3137.5152,1.6384,9.12,1,10_50,,,,0.4615,0.2308,0.3077,0,0.462,0.231,0.308,0,,,,,,,,,,,
+,0.431,0.231,0.315,0.023,0,0,8.731,137.27,124.569,1.8654,8419.162,89.163,1515.239,753.4459,1634.423,3145.7622,1.6234,9.78,1,10_50,,,,0.4308,0.2308,0.3385,0,0.431,0.231,0.338,0,,,,,,,,,,,
+,0.408,0.231,0.338,0.023,0,0,8.754,137.155,124.592,1.8665,8442.76,89.623,1528.74,753.3983,1636.3875,3152.0849,1.6119,10.65,1,10_50,,,,0.3846,0.2308,0.3846,0,0.385,0.231,0.385,0,,,,,,,,,,,
+,0.646,0.231,0.077,0.046,0,0,8.447,138.46,124.354,1.8546,8158.092,84.725,1398.51,751.4178,1614.7328,3081.0191,1.6776,15.52,1,10_50,,,,0.3462,0.2308,0.4231,0,0.346,0.231,0.423,0,,,,,,,,,,,
+,0.623,0.231,0.1,0.046,0,0,8.47,138.345,124.377,1.8558,8181.69,85.185,1412.011,751.3702,1616.6972,3087.3418,1.6661,12.56,1,10_50,,,,0.6308,0.2308,0.1154,0.0231,0.631,0.231,0.115,0.023,,,,,,,,,,,
+,0.592,0.231,0.131,0.046,0,0,8.501,138.19,124.408,1.8573,8213.496,85.805,1430.208,751.306,1619.3449,3095.8637,1.6506,3.83,1,10_50,,,,0.5923,0.2308,0.1539,0.0231,0.592,0.231,0.154,0.023,,,,,,,,,,,
+,0.549,0.231,0.174,0.046,0,0,8.524,138.075,124.431,1.8585,8237.094,86.265,1443.709,751.2584,1621.3094,3102.1864,1.6391,2.54,1,10_50,,,,0.5539,0.2308,0.1923,0.0231,0.554,0.231,0.192,0.023,,,,,,,,,,,
+,0.561,0.231,0.162,0.046,0,0,8.532,138.035,124.439,1.8589,8245.302,86.425,1448.405,751.2418,1621.9926,3104.3856,1.6351,1.87,1,10_50,,,,0.5,0.2308,0.2462,0.0231,0.5,0.231,0.246,0.023,,,,,,,,,,,
+,0.531,0.231,0.192,0.046,0,0,8.562,137.885,124.469,1.8604,8276.082,87.025,1466.015,751.1797,1624.5549,3112.6326,1.6201,3.63,1,10_50,,,,0.4615,0.2308,0.2846,0.0231,0.462,0.231,0.285,0.023,,,,,,,,,,,
+,0.5,0.231,0.223,0.046,0,0,8.593,137.73,124.5,1.8619,8307.888,87.645,1484.212,751.1156,1627.2026,3121.1545,1.6046,4.75,1,10_50,,,,0.4308,0.2308,0.3154,0.0231,0.431,0.231,0.315,0.023,,,,,,,,,,,
+,0.461,0.231,0.262,0.046,0,0,8.632,137.535,124.539,1.8639,8347.902,88.425,1507.105,751.0348,1630.5336,3131.8756,1.5851,6.14,1,10_50,,,,0.4077,0.2308,0.3385,0.0231,0.408,0.231,0.338,0.023,,,,,,,,,,,
+,0.431,0.231,0.302,0.056,0,0,8.662,137.385,124.569,1.8654,8378.682,89.025,1524.715,750.9727,1633.0959,3140.1226,1.5701,8.77,1,10_50,,,,0.6462,0.2308,0.0769,0.0462,0.646,0.231,0.077,0.046,,,,,,,,,,,
+,0.385,0.231,0.338,0.046,0,0,8.708,137.155,124.615,1.8677,8425.878,89.945,1551.717,750.8775,1637.0248,3152.768,1.5471,10.4,1,10_50,,,,0.6231,0.2308,0.1,0.0462,0.623,0.231,0.1,0.046,,,,,,,,,,,
+,0.346,0.231,0.377,0.046,0,0,8.747,136.96,124.654,1.8696,8465.892,90.725,1574.61,750.7968,1640.3558,3163.4891,1.5276,11.02,1,10_50,,,,0.5923,0.2308,0.1308,0.0462,0.592,0.231,0.131,0.046,,,,,,,,,,,
+,0.569,0.231,0.138,0.062,0,0,8.476,138.155,124.431,1.8585,8208.934,86.169,1450.301,749.5379,1620.3862,3098.2632,1.6019,4.18,1,10_50,,,,0.5692,0.2308,0.1539,0.0462,0.569,0.231,0.154,0.046,,,,,,,,,,,
+,0.607,0.231,0.085,0.077,0,0,8.393,138.42,124.393,1.8566,8143.546,85.319,1434.175,748.0036,1616.2751,3084.139,1.5861,15.15,1,10_50,,,,0.5615,0.2308,0.1615,0.0462,0.562,0.231,0.162,0.046,,,,,,,,,,,
+,0.567,0.231,0.125,0.077,0,0,8.423,138.27,124.423,1.8581,8174.326,85.919,1451.785,747.9415,1618.8374,3092.386,1.5711,10.65,1,10_50,,,,0.5308,0.2308,0.1923,0.0462,0.531,0.231,0.192,0.046,,,,,,,,,,,
+,0.546,0.231,0.146,0.077,0,0,8.454,138.115,124.454,1.8596,8206.132,86.539,1469.982,747.8774,1621.4851,3100.9079,1.5556,5.31,1,10_50,,,,0.5,0.2308,0.2231,0.0462,0.5,0.231,0.223,0.046,,,,,,,,,,,
+,0.5,0.231,0.192,0.077,0,0,8.5,137.885,124.5,1.8619,8253.328,87.459,1496.984,747.7821,1625.4139,3113.5533,1.5326,4.9,1,10_50,,,,0.4615,0.2308,0.2615,0.0462,0.462,0.231,0.262,0.046,,,,,,,,,,,
+,0.461,0.231,0.231,0.077,0,0,8.539,137.69,124.539,1.8639,8293.342,88.239,1519.877,747.7014,1628.7449,3124.2744,1.5131,6.47,1,10_50,,,,0.4308,0.2308,0.2923,0.0462,0.431,0.231,0.292,0.046,,,,,,,,,,,
+,0.43,0.231,0.262,0.077,0,0,8.57,137.535,124.57,1.8654,8325.148,88.859,1538.074,747.6372,1631.3926,3132.7963,1.4976,7.5,1,10_50,,,,0.3846,0.2308,0.3385,0.0462,0.385,0.231,0.338,0.046,,,,,,,,,,,
+,0.394,0.231,0.318,0.077,0,0,8.616,137.305,124.616,1.8677,8372.344,89.779,1565.076,747.542,1635.3215,3145.4417,1.4746,9.26,1,10_50,,,,0.3462,0.2308,0.3769,0.0462,0.346,0.231,0.377,0.046,,,,,,,,,,,
+,0.346,0.231,0.346,0.077,0,0,8.654,137.115,124.654,1.8696,8411.332,90.539,1587.382,747.4634,1638.5671,3155.8879,1.4556,9.96,1,10_50,,,,0.5692,0.2308,0.1385,0.0615,0.569,0.231,0.138,0.062,,,,,,,,,,,
+,0.307,0.231,0.385,0.077,0,0,8.693,136.92,124.693,1.8716,8451.346,91.319,1610.275,747.3826,1641.8981,3166.609,1.4361,11.39,1,10_50,,,,0.6077,0.2308,0.0846,0.0769,0.608,0.231,0.085,0.077,,,,,,,,,,,
+,0.392,0.231,0.277,0.1,0,0,8.539,137.46,124.608,1.8673,8323.656,89.481,1569.856,745.0854,1633.3111,3137.6029,1.4253,8.1,1,10_50,,,,0.5769,0.2308,0.1154,0.0769,0.577,0.231,0.115,0.077,,,,,,,,,,,
+,0.561,0.231,0.1,0.108,0,0,8.346,138.345,124.439,1.8589,8136.182,86.053,1473.949,744.575,1618.4152,3089.1832,1.4912,13.7,1,10_50,,,,0.5462,0.2308,0.1462,0.0769,0.546,0.231,0.146,0.077,,,,,,,,,,,
+,0.53,0.231,0.131,0.108,0,0,8.377,138.19,124.47,1.8604,8167.988,86.673,1492.146,744.5108,1621.0629,3097.7051,1.4757,11.12,1,10_50,,,,0.5,0.2308,0.1923,0.0769,0.5,0.231,0.192,0.077,,,,,,,,,,,
+,0.489,0.231,0.172,0.108,0,0,8.408,138.035,124.501,1.862,8199.794,87.293,1510.343,744.4466,1623.7107,3106.227,1.4602,8.2,1,10_50,,,,0.4615,0.2308,0.2308,0.0769,0.462,0.231,0.231,0.077,,,,,,,,,,,
+,0.461,0.231,0.2,0.108,0,0,8.446,137.845,124.539,1.8639,8238.782,88.053,1532.649,744.368,1626.9562,3116.6732,1.4412,6.4,1,10_50,,,,0.4308,0.2308,0.2615,0.0769,0.431,0.231,0.262,0.077,,,,,,,,,,,
+,0.43,0.231,0.231,0.108,0,0,8.477,137.69,124.57,1.8654,8270.588,88.673,1550.846,744.3038,1629.6039,3125.1951,1.4257,7.19,1,10_50,,,,0.3846,0.2308,0.3077,0.0769,0.385,0.231,0.308,0.077,,,,,,,,,,,
+,0.346,0.231,0.315,0.108,0,0,8.561,137.27,124.654,1.8696,8356.772,90.353,1600.154,744.1299,1636.7784,3148.2867,1.3837,9.04,1,10_50,,,,0.3462,0.2308,0.3462,0.0769,0.346,0.231,0.346,0.077,,,,,,,,,,,
+,0.315,0.231,0.346,0.108,0,0,8.592,137.115,124.685,1.8712,8388.578,90.973,1618.351,744.0658,1639.4261,3156.8086,1.3682,9.08,1,10_50,,,,0.3077,0.2308,0.3846,0.0769,0.308,0.231,0.385,0.077,,,,,,,,,,,
+,0.292,0.231,0.369,0.108,0,0,8.615,137,124.708,1.8723,8412.176,91.433,1631.852,744.0181,1641.3905,3163.1313,1.3567,9.11,1,10_50,,,,0.3923,0.2308,0.2769,0.1,0.392,0.231,0.277,0.1,,,,,,,,,,,
+,0.384,0.231,0.254,0.131,0,0,8.454,137.575,124.616,1.8677,8277.304,89.455,1587.324,741.7354,1632.2057,3132.2009,1.3494,9.33,1,10_50,,,,0.5615,0.2308,0.1,0.1077,0.562,0.231,0.1,0.108,,,,,,,,,,,
+,0.346,0.231,0.292,0.131,0,0,8.492,137.385,124.654,1.8696,8316.292,90.215,1609.63,741.6567,1635.4513,3142.6471,1.3304,9.1,1,10_50,,,,0.5308,0.2308,0.1308,0.1077,0.531,0.231,0.131,0.108,,,,,,,,,,,
+,0.538,0.231,0.077,0.154,0,0,8.231,138.46,124.462,1.86,8078.82,86.237,1506.402,739.581,1617.7255,3084.2267,1.373,16.6,1,10_50,,,,0.5,0.2308,0.1615,0.1077,0.5,0.231,0.162,0.108,,,,,,,,,,,
+,0.507,0.231,0.108,0.154,0,0,8.262,138.305,124.493,1.8616,8110.626,86.857,1524.599,739.5168,1620.3732,3092.7486,1.3575,15.39,1,10_50,,,,0.4615,0.2308,0.2,0.1077,0.462,0.231,0.2,0.108,,,,,,,,,,,
+,0.461,0.231,0.154,0.154,0,0,8.308,138.075,124.539,1.8639,8157.822,87.777,1551.601,739.4216,1624.302,3105.394,1.3345,15.2,1,10_50,,,,0.4308,0.2308,0.2308,0.1077,0.431,0.231,0.231,0.108,,,,,,,,,,,
+,0.43,0.231,0.185,0.154,0,0,8.339,137.92,124.57,1.8654,8189.628,88.397,1569.798,739.3574,1626.9497,3113.9159,1.319,14.01,1,10_50,,,,0.3462,0.2308,0.3154,0.1077,0.346,0.231,0.315,0.108,,,,,,,,,,,
+,0.384,0.231,0.231,0.154,0,0,8.385,137.69,124.616,1.8677,8236.824,89.317,1596.8,739.2622,1630.8786,3126.5613,1.296,11.57,1,10_50,,,,0.3154,0.2308,0.3462,0.1077,0.315,0.231,0.346,0.108,,,,,,,,,,,
+,0.346,0.231,0.269,0.154,0,0,8.423,137.5,124.654,1.8696,8275.812,90.077,1619.106,739.1835,1634.1242,3137.0075,1.277,10.32,1,10_50,,,,0.2923,0.2308,0.3692,0.1077,0.292,0.231,0.369,0.108,,,,,,,,,,,
+,0.269,0.231,0.346,0.154,0,0,8.5,137.115,124.731,1.8735,8354.814,91.617,1664.305,739.0242,1640.7008,3158.1748,1.2385,10.52,1,10_50,,,,0.3846,0.2308,0.2539,0.1308,0.385,0.231,0.254,0.131,,,,,,,,,,,
+,0.714,0.286,0,0,0,0,8.572,138.57,124.286,1.8529,8169.724,83.146,1337.442,755.2256,1617.3343,3082.5536,1.7567,13.61,1,10_50,,,,0.3462,0.2308,0.2923,0.1308,0.346,0.231,0.292,0.131,,,,,,,,,,,
+,0.685,0.286,0.029,0,0,0,8.601,138.425,124.315,1.8543,8199.478,83.726,1354.465,755.1656,1619.8112,3090.5257,1.7422,9.11,1,10_50,,,,0.5385,0.2308,0.0769,0.1539,0.538,0.231,0.077,0.154,,,,,,,,,,,
+,0.635,0.286,0.079,0,0,0,8.651,138.175,124.365,1.8568,8250.778,84.726,1383.815,755.0621,1624.0817,3104.2707,1.7172,8.02,1,10_50,,,,0.5077,0.2308,0.1077,0.1539,0.508,0.231,0.108,0.154,,,,,,,,,,,
+,0.607,0.286,0.107,0,0,0,8.679,138.035,124.393,1.8582,8279.506,85.286,1400.251,755.0041,1626.4732,3111.9679,1.7032,8.36,1,10_50,,,,0.4615,0.2308,0.1539,0.1539,0.462,0.231,0.154,0.154,,,,,,,,,,,
+,0.571,0.286,0.143,0,0,0,8.715,137.855,124.429,1.86,8316.442,86.006,1421.383,754.9296,1629.5479,3121.8643,1.6852,4.47,1,10_50,,,,0.4308,0.2308,0.1846,0.1539,0.431,0.231,0.185,0.154,,,,,,,,,,,
+,0.535,0.286,0.179,0,0,0,8.751,137.675,124.465,1.8618,8353.378,86.726,1442.515,754.8551,1632.6227,3131.7607,1.6672,6.6,1,10_50,,,,0.3846,0.2308,0.2308,0.1539,0.385,0.231,0.231,0.154,,,,,,,,,,,
+,0.5,0.286,0.214,0,0,0,8.786,137.5,124.5,1.8636,8389.288,87.426,1463.06,754.7826,1635.612,3141.3822,1.6497,8.74,1,10_50,,,,0.3462,0.2308,0.2692,0.1539,0.346,0.231,0.269,0.154,,,,,,,,,,,
+,0.464,0.286,0.25,0,0,0,8.822,137.32,124.536,1.8654,8426.224,88.146,1484.192,754.7081,1638.6868,3151.2786,1.6317,9.97,1,10_50,,,,0.2692,0.2308,0.3462,0.1539,0.269,0.231,0.346,0.154,,,,,,,,,,,
+,0.428,0.286,0.286,0,0,0,8.858,137.14,124.572,1.8672,8463.16,88.866,1505.324,754.6336,1641.7616,3161.175,1.6137,10.84,1,10_50,,,,0.7143,0.2857,0,0,0.714,0.286,0,0,,,,,,,,,,,
+,0.393,0.286,0.321,0,0,0,8.893,136.965,124.607,1.8689,8499.07,89.566,1525.869,754.5612,1644.7509,3170.7965,1.5962,12.22,1,10_50,,,,0.6857,0.2857,0.0286,0,0.686,0.286,0.029,0,,,,,,,,,,,
+,0.25,0.286,0.464,0,0,0,9.036,136.25,124.75,1.8761,8645.788,92.426,1609.81,754.2651,1656.9645,3210.1072,1.5247,12.07,1,10_50,,,,0.6357,0.2857,0.0786,0,0.636,0.286,0.079,0,,,,,,,,,,,
+,0.69,0.286,0.01,0.014,0,0,8.544,138.57,124.3,1.8536,8159.448,83.342,1351.428,753.6912,1617.7222,3082.9694,1.7172,14.55,1,10_50,,,,0.6071,0.2857,0.1071,0,0.607,0.286,0.107,0,,,,,,,,,,,
+,0.686,0.286,0.014,0.014,0,0,8.558,138.5,124.314,1.8543,8173.812,83.622,1359.646,753.6622,1618.918,3086.818,1.7102,9.32,1,10_50,,,,0.5714,0.2857,0.1429,0,0.571,0.286,0.143,0,,,,,,,,,,,
+,0.65,0.286,0.05,0.014,0,0,8.594,138.32,124.35,1.8561,8210.748,84.342,1380.778,753.5877,1621.9927,3096.7144,1.6922,3.05,1,10_50,,,,0.5357,0.2857,0.1786,0,0.536,0.286,0.179,0,,,,,,,,,,,
+,0.636,0.286,0.064,0.014,0,0,8.608,138.25,124.364,1.8568,8225.112,84.622,1388.996,753.5587,1623.1885,3100.563,1.6852,1.56,1,10_50,,,,0.5,0.2857,0.2143,0,0.5,0.286,0.214,0,,,,,,,,,,,
+,0.621,0.286,0.079,0.014,0,0,8.623,138.175,124.379,1.8575,8240.502,84.922,1397.801,753.5277,1624.4696,3104.6865,1.6777,0.54,1,10_50,,,,0.4643,0.2857,0.25,0,0.464,0.286,0.25,0,,,,,,,,,,,
+,0.607,0.286,0.093,0.014,0,0,8.637,138.105,124.393,1.8582,8254.866,85.202,1406.019,753.4987,1625.6654,3108.5351,1.6707,1.04,1,10_50,,,,0.4286,0.2857,0.2857,0,0.429,0.286,0.286,0,,,,,,,,,,,
+,0.593,0.286,0.107,0.014,0,0,8.651,138.035,124.407,1.8589,8269.23,85.482,1414.237,753.4697,1626.8611,3112.3837,1.6637,1.37,1,10_50,,,,0.3929,0.2857,0.3214,0,0.393,0.286,0.321,0,,,,,,,,,,,
+,0.579,0.286,0.121,0.014,0,0,8.665,137.965,124.421,1.8596,8283.594,85.762,1422.455,753.4408,1628.0569,3116.2323,1.6567,2.53,1,10_50,,,,0.25,0.2857,0.4643,0,0.25,0.286,0.464,0,,,,,,,,,,,
+,0.685,0.286,0,0.029,0,0,8.514,138.57,124.315,1.8543,8148.438,83.552,1366.413,752.0472,1618.1379,3083.4149,1.6749,13.77,1,10_50,,,,0.7,0.2857,0,0.0143,0.7,0.286,0,0.014,,,,,,,,,,,
+,0.664,0.286,0.021,0.029,0,0,8.535,138.465,124.336,1.8554,8169.984,83.972,1378.74,752.0038,1619.9315,3089.1878,1.6644,9.7,1,10_50,,,,0.6857,0.2857,0.0143,0.0143,0.686,0.286,0.014,0.014,,,,,,,,,,,
+,0.635,0.286,0.05,0.029,0,0,8.564,138.32,124.365,1.8568,8199.738,84.552,1395.763,751.9437,1622.4084,3097.1599,1.6499,4,1,10_50,,,,0.65,0.2857,0.05,0.0143,0.65,0.286,0.05,0.014,,,,,,,,,,,
+,0.606,0.286,0.079,0.029,0,0,8.593,138.175,124.394,1.8583,8229.492,85.132,1412.786,751.8837,1624.8853,3105.132,1.6354,1.84,1,10_50,,,,0.6357,0.2857,0.0643,0.0143,0.636,0.286,0.064,0.014,,,,,,,,,,,
+,0.592,0.286,0.093,0.029,0,0,8.607,138.105,124.408,1.859,8243.856,85.412,1421.004,751.8547,1626.081,3108.9806,1.6284,1.58,1,10_50,,,,0.6214,0.2857,0.0786,0.0143,0.621,0.286,0.079,0.014,,,,,,,,,,,
+,0.578,0.286,0.107,0.029,0,0,8.621,138.035,124.422,1.8597,8258.22,85.692,1429.222,751.8257,1627.2768,3112.8292,1.6214,2.26,1,10_50,,,,0.6071,0.2857,0.0929,0.0143,0.607,0.286,0.093,0.014,,,,,,,,,,,
+,0.564,0.286,0.121,0.029,0,0,8.635,137.965,124.436,1.8604,8272.584,85.972,1437.44,751.7968,1628.4725,3116.6778,1.6144,2.92,1,10_50,,,,0.5929,0.2857,0.1071,0.0143,0.593,0.286,0.107,0.014,,,,,,,,,,,
+,0.542,0.286,0.143,0.029,0,0,8.657,137.855,124.458,1.8615,8295.156,86.412,1450.354,751.7512,1630.3515,3122.7256,1.6034,4.29,1,10_50,,,,0.5786,0.2857,0.1214,0.0143,0.579,0.286,0.121,0.014,,,,,,,,,,,
+,0.521,0.286,0.164,0.029,0,0,8.678,137.75,124.479,1.8625,8316.702,86.832,1462.681,751.7077,1632.1451,3128.4985,1.5929,5.98,1,10_50,,,,0.6857,0.2857,0,0.0286,0.686,0.286,0,0.029,,,,,,,,,,,
+,0.499,0.286,0.186,0.029,0,0,8.7,137.64,124.501,1.8636,8339.274,87.272,1475.595,751.6622,1634.0242,3134.5463,1.5819,7.53,1,10_50,,,,0.6643,0.2857,0.0214,0.0286,0.664,0.286,0.021,0.029,,,,,,,,,,,
+,0.464,0.286,0.221,0.029,0,0,8.735,137.465,124.536,1.8654,8375.184,87.972,1496.14,751.5898,1637.0135,3144.1678,1.5644,8.91,1,10_50,,,,0.6357,0.2857,0.05,0.0286,0.636,0.286,0.05,0.029,,,,,,,,,,,
+,0.448,0.286,0.237,0.029,0,0,8.771,137.285,124.572,1.8672,8412.12,88.692,1517.272,751.5152,1640.0883,3154.0642,1.5464,10.09,1,10_50,,,,0.6071,0.2857,0.0786,0.0286,0.607,0.286,0.079,0.029,,,,,,,,,,,
+,0.378,0.286,0.307,0.029,0,0,8.821,137.035,124.622,1.8697,8463.42,89.692,1546.622,751.4117,1644.3588,3167.8092,1.5214,11.34,1,10_50,,,,0.5929,0.2857,0.0929,0.0286,0.593,0.286,0.093,0.029,,,,,,,,,,,
+,0.635,0.286,0.036,0.043,0,0,8.522,138.39,124.365,1.8568,8175.098,84.468,1401.531,750.4383,1621.6006,3093.7271,1.6174,7.18,1,10_50,,,,0.5786,0.2857,0.1071,0.0286,0.579,0.286,0.107,0.029,,,,,,,,,,,
+,0.614,0.286,0.057,0.043,0,0,8.543,138.285,124.386,1.8579,8196.644,84.888,1413.858,750.3948,1623.3942,3099.5,1.6069,5.22,1,10_50,,,,0.5643,0.2857,0.1214,0.0286,0.564,0.286,0.121,0.029,,,,,,,,,,,
+,0.592,0.286,0.079,0.043,0,0,8.565,138.175,124.408,1.859,8219.216,85.328,1426.772,750.3493,1625.2732,3105.5478,1.5959,2.79,1,10_50,,,,0.5429,0.2857,0.1429,0.0286,0.543,0.286,0.143,0.029,,,,,,,,,,,
+,0.578,0.286,0.093,0.043,0,0,8.579,138.105,124.422,1.8597,8233.58,85.608,1434.99,750.3203,1626.469,3109.3964,1.5889,2.98,1,10_50,,,,0.5214,0.2857,0.1643,0.0286,0.521,0.286,0.164,0.029,,,,,,,,,,,
+,0.564,0.286,0.107,0.043,0,0,8.593,138.035,124.436,1.8604,8247.944,85.888,1443.208,750.2913,1627.6647,3113.245,1.5819,3.08,1,10_50,,,,0.5,0.2857,0.1857,0.0286,0.5,0.286,0.186,0.029,,,,,,,,,,,
+,0.55,0.286,0.121,0.043,0,0,8.607,137.965,124.45,1.8611,8262.308,86.168,1451.426,750.2624,1628.8604,3117.0936,1.5749,3.49,1,10_50,,,,0.4643,0.2857,0.2214,0.0286,0.464,0.286,0.221,0.029,,,,,,,,,,,
+,0.664,0.286,0,0.05,0,0,8.472,138.57,124.336,1.8554,8133.024,83.846,1387.392,749.7456,1618.7198,3084.0386,1.6157,14.23,1,10_50,,,,0.4286,0.2857,0.2571,0.0286,0.429,0.286,0.257,0.029,,,,,,,,,,,
+,0.636,0.286,0.021,0.057,0,0,8.479,138.465,124.364,1.8568,8149.432,84.364,1406.712,748.935,1620.7074,3090.0194,1.5854,11.32,1,10_50,,,,0.3786,0.2857,0.3071,0.0286,0.379,0.286,0.307,0.029,,,,,,,,,,,
+,0.593,0.286,0.064,0.057,0,0,8.522,138.25,124.407,1.8589,8193.55,85.224,1431.953,748.8459,1624.38,3101.8401,1.5639,6.14,1,10_50,,,,0.6357,0.2857,0.0357,0.0429,0.636,0.286,0.036,0.043,,,,,,,,,,,
+,0.578,0.286,0.079,0.057,0,0,8.537,138.175,124.422,1.8597,8208.94,85.524,1440.758,748.8149,1625.6612,3105.9636,1.5564,4.69,1,10_50,,,,0.6143,0.2857,0.0571,0.0429,0.614,0.286,0.057,0.043,,,,,,,,,,,
+,0.564,0.286,0.093,0.057,0,0,8.551,138.105,124.436,1.8604,8223.304,85.804,1448.976,748.7859,1626.8569,3109.8122,1.5494,4.03,1,10_50,,,,0.5929,0.2857,0.0786,0.0429,0.593,0.286,0.079,0.043,,,,,,,,,,,
+,0.55,0.286,0.107,0.057,0,0,8.565,138.035,124.45,1.8611,8237.668,86.084,1457.194,748.7569,1628.0526,3113.6608,1.5424,4.21,1,10_50,,,,0.5786,0.2857,0.0929,0.0429,0.579,0.286,0.093,0.043,,,,,,,,,,,
+,0.536,0.286,0.121,0.057,0,0,8.579,137.965,124.464,1.8618,8252.032,86.364,1465.412,748.728,1629.2484,3117.5094,1.5354,4.12,1,10_50,,,,0.5643,0.2857,0.1071,0.0429,0.564,0.286,0.107,0.043,,,,,,,,,,,
+,0.514,0.286,0.143,0.057,0,0,8.601,137.855,124.486,1.8629,8274.604,86.804,1478.326,748.6824,1631.1274,3123.5572,1.5244,5.5,1,10_50,,,,0.55,0.2857,0.1214,0.0429,0.55,0.286,0.121,0.043,,,,,,,,,,,
+,0.643,0.286,0,0.071,0,0,8.43,138.57,124.357,1.8564,8117.61,84.14,1408.371,747.444,1619.3017,3084.6623,1.5565,14.88,1,10_50,,,,0.6643,0.2857,0,0.05,0.664,0.286,0,0.05,,,,,,,,,,,
+,0.622,0.286,0.021,0.071,0,0,8.451,138.465,124.378,1.8575,8139.156,84.56,1420.698,747.4006,1621.0953,3090.4352,1.546,12.81,1,10_50,,,,0.6357,0.2857,0.0214,0.0571,0.636,0.286,0.021,0.057,,,,,,,,,,,
+,0.593,0.286,0.05,0.071,0,0,8.48,138.32,124.407,1.8589,8168.91,85.14,1437.721,747.3405,1623.5722,3098.4073,1.5315,7.55,1,10_50,,,,0.5929,0.2857,0.0643,0.0571,0.593,0.286,0.064,0.057,,,,,,,,,,,
+,0.579,0.286,0.064,0.071,0,0,8.494,138.25,124.421,1.8596,8183.274,85.42,1445.939,747.3115,1624.768,3102.2559,1.5245,6.09,1,10_50,,,,0.5786,0.2857,0.0786,0.0571,0.579,0.286,0.079,0.057,,,,,,,,,,,
+,0.584,0.286,0.059,0.071,0,0,8.509,138.175,124.436,1.8604,8198.664,85.72,1454.744,747.2805,1626.0491,3106.3794,1.517,4.97,1,10_50,,,,0.5643,0.2857,0.0929,0.0571,0.564,0.286,0.093,0.057,,,,,,,,,,,
+,0.55,0.286,0.093,0.071,0,0,8.523,138.105,124.45,1.8611,8213.028,86,1462.962,747.2515,1627.2448,3110.228,1.51,5.47,1,10_50,,,,0.55,0.2857,0.1071,0.0571,0.55,0.286,0.107,0.057,,,,,,,,,,,
+,0.536,0.286,0.107,0.071,0,0,8.537,138.035,124.464,1.8618,8227.392,86.28,1471.18,747.2225,1628.4406,3114.0766,1.503,4.58,1,10_50,,,,0.5357,0.2857,0.1214,0.0571,0.536,0.286,0.121,0.057,,,,,,,,,,,
+,0.5,0.286,0.143,0.071,0,0,8.573,137.855,124.5,1.8636,8264.328,87,1492.312,747.148,1631.5153,3123.973,1.485,5.35,1,10_50,,,,0.5143,0.2857,0.1429,0.0571,0.514,0.286,0.143,0.057,,,,,,,,,,,
+,0.464,0.286,0.179,0.071,0,0,8.609,137.675,124.536,1.8654,8301.264,87.72,1513.444,747.0735,1634.5901,3133.8694,1.467,7.01,1,10_50,,,,0.6429,0.2857,0,0.0714,0.643,0.286,0,0.071,,,,,,,,,,,
+,0.429,0.286,0.214,0.071,0,0,8.644,137.5,124.571,1.8671,8337.174,88.42,1533.989,747.001,1637.5795,3143.4909,1.4495,9,1,10_50,,,,0.6214,0.2857,0.0214,0.0714,0.621,0.286,0.021,0.071,,,,,,,,,,,
+,0.357,0.286,0.286,0.071,0,0,8.716,137.14,124.643,1.8707,8411.046,89.86,1576.253,746.852,1643.729,3163.2837,1.4135,10.83,1,10_50,,,,0.5929,0.2857,0.05,0.0714,0.593,0.286,0.05,0.071,,,,,,,,,,,
+,0.606,0.286,0.029,0.079,0,0,8.443,138.425,124.394,1.8583,8141.492,84.832,1433.386,746.5072,1622.0003,3092.872,1.5194,9.67,1,10_50,,,,0.5786,0.2857,0.0643,0.0714,0.579,0.286,0.064,0.071,,,,,,,,,,,
+,0.621,0.286,0,0.093,0,0,8.386,138.57,124.379,1.8575,8101.462,84.448,1430.349,745.0328,1619.9113,3085.3157,1.4944,15.21,1,10_50,,,,0.5643,0.2857,0.0786,0.0714,0.564,0.286,0.079,0.071,,,,,,,,,,,
+,0.571,0.286,0.05,0.093,0,0,8.436,138.32,124.429,1.86,8152.762,85.448,1459.699,744.9293,1624.1818,3099.0607,1.4694,9.53,1,10_50,,,,0.55,0.2857,0.0929,0.0714,0.55,0.286,0.093,0.071,,,,,,,,,,,
+,0.55,0.286,0.071,0.093,0,0,8.457,138.215,124.45,1.8611,8174.308,85.868,1472.026,744.8859,1625.9754,3104.8336,1.4589,8.86,1,10_50,,,,0.5357,0.2857,0.1071,0.0714,0.536,0.286,0.107,0.071,,,,,,,,,,,
+,0.535,0.286,0.086,0.093,0,0,8.472,138.14,124.465,1.8618,8189.698,86.168,1480.831,744.8548,1627.2566,3108.9571,1.4514,7.64,1,10_50,,,,0.5,0.2857,0.1429,0.0714,0.5,0.286,0.143,0.071,,,,,,,,,,,
+,0.593,0.286,0.014,0.107,0,0,8.372,138.5,124.407,1.8589,8105.55,84.924,1452.553,743.4694,1621.495,3089.5801,1.4479,14.3,1,10_50,,,,0.4643,0.2857,0.1786,0.0714,0.464,0.286,0.179,0.071,,,,,,,,,,,
+,0.578,0.286,0.029,0.107,0,0,8.387,138.425,124.422,1.8597,8120.94,85.224,1461.358,743.4384,1622.7762,3093.7036,1.4404,13.12,1,10_50,,,,0.4286,0.2857,0.2143,0.0714,0.429,0.286,0.214,0.071,,,,,,,,,,,
+,0.55,0.286,0.057,0.107,0,0,8.415,138.285,124.45,1.8611,8149.668,85.784,1477.794,743.3804,1625.1676,3101.4008,1.4264,12.94,1,10_50,,,,0.3571,0.2857,0.2857,0.0714,0.357,0.286,0.286,0.071,,,,,,,,,,,
+,0.5,0.286,0.107,0.107,0,0,8.465,138.035,124.5,1.8636,8200.968,86.784,1507.144,743.2769,1629.4381,3115.1458,1.4014,7.62,1,10_50,,,,0.6071,0.2857,0.0286,0.0786,0.607,0.286,0.029,0.079,,,,,,,,,,,
+,0.464,0.286,0.143,0.107,0,0,8.501,137.855,124.536,1.8654,8237.904,87.504,1528.276,743.2024,1632.5129,3125.0422,1.3834,8.09,1,10_50,,,,0.6214,0.2857,0,0.0929,0.621,0.286,0,0.093,,,,,,,,,,,
+,0.428,0.286,0.179,0.107,0,0,8.537,137.675,124.572,1.8672,8274.84,88.224,1549.408,743.1279,1635.5877,3134.9386,1.3654,8.31,1,10_50,,,,0.5714,0.2857,0.05,0.0929,0.571,0.286,0.05,0.093,,,,,,,,,,,
+,0.393,0.286,0.214,0.107,0,0,8.572,137.5,124.607,1.8689,8310.75,88.924,1569.953,743.0554,1638.577,3144.5601,1.3479,8.97,1,10_50,,,,0.55,0.2857,0.0714,0.0929,0.55,0.286,0.071,0.093,,,,,,,,,,,
+,0.357,0.286,0.25,0.107,0,0,8.608,137.32,124.643,1.8707,8347.686,89.644,1591.085,742.9809,1641.6518,3154.4565,1.3299,10.44,1,10_50,,,,0.5357,0.2857,0.0857,0.0929,0.536,0.286,0.086,0.093,,,,,,,,,,,
+,0.571,0.286,0,0.143,0,0,8.286,138.57,124.429,1.86,8064.762,85.148,1480.299,739.5528,1621.2968,3086.8007,1.3534,16.57,1,10_50,,,,0.5929,0.2857,0.0143,0.1071,0.593,0.286,0.014,0.107,,,,,,,,,,,
+,0.535,0.286,0.036,0.143,0,0,8.322,138.39,124.465,1.8618,8101.698,85.868,1501.431,739.4783,1624.3716,3096.6971,1.3354,15.75,1,10_50,,,,0.5786,0.2857,0.0286,0.1071,0.579,0.286,0.029,0.107,,,,,,,,,,,
+,0.5,0.286,0.071,0.143,0,0,8.357,138.215,124.5,1.8636,8137.608,86.568,1521.976,739.4059,1627.3609,3106.3186,1.3179,13.19,1,10_50,,,,0.55,0.2857,0.0571,0.1071,0.55,0.286,0.057,0.107,,,,,,,,,,,
+,0.464,0.286,0.107,0.143,0,0,8.393,138.035,124.536,1.8654,8174.544,87.288,1543.108,739.3313,1630.4357,3116.215,1.2999,11.78,1,10_50,,,,0.5,0.2857,0.1071,0.1071,0.5,0.286,0.107,0.107,,,,,,,,,,,
+,0.428,0.286,0.143,0.143,0,0,8.429,137.855,124.572,1.8672,8211.48,88.008,1564.24,739.2568,1633.5105,3126.1114,1.2819,10.66,1,10_50,,,,0.4643,0.2857,0.1429,0.1071,0.464,0.286,0.143,0.107,,,,,,,,,,,
+,0.392,0.286,0.179,0.143,0,0,8.465,137.675,124.608,1.869,8248.416,88.728,1585.372,739.1823,1636.5852,3136.0078,1.2639,9.96,1,10_50,,,,0.4286,0.2857,0.1786,0.1071,0.429,0.286,0.179,0.107,,,,,,,,,,,
+,0.357,0.286,0.214,0.143,0,0,8.5,137.5,124.643,1.8707,8284.326,89.428,1605.917,739.1098,1639.5746,3145.6293,1.2464,10.08,1,10_50,,,,0.3929,0.2857,0.2143,0.1071,0.393,0.286,0.214,0.107,,,,,,,,,,,
+,0.321,0.286,0.25,0.143,0,0,8.536,137.32,124.679,1.8725,8321.262,90.148,1627.049,739.0353,1642.6493,3155.5257,1.2284,9.63,1,10_50,,,,0.3571,0.2857,0.25,0.1071,0.357,0.286,0.25,0.107,,,,,,,,,,,
+Fe-Co-Ni-V,0.618,0.091,0.291,0,0,0,8.473,138.09,124.382,1.8518,8266.66,86.821,1401.594,759.5626,1605.1289,3077.2175,1.9271,10.53,1,10_50,,,,0.5714,0.2857,0,0.1429,0.571,0.286,0,0.143,,,,,,,,,,,
+,0.527,0.091,0.382,0,0,0,8.564,137.635,124.473,1.8564,8360.026,88.641,1455.011,759.3742,1612.9012,3102.2334,1.8816,10.57,1,10_50,,,,0.5357,0.2857,0.0357,0.1429,0.536,0.286,0.036,0.143,,,,,,,,,,,
+,0.473,0.091,0.436,0,0,0,8.618,137.365,124.527,1.8591,8415.43,89.721,1486.709,759.2625,1617.5133,3117.078,1.8546,11.05,1,10_50,,,,0.5,0.2857,0.0714,0.1429,0.5,0.286,0.071,0.143,,,,,,,,,,,
+,0.4,0.091,0.509,0,0,0,8.691,137,124.6,1.8627,8490.328,91.181,1529.56,759.1113,1623.7482,3137.1457,1.8181,11.06,1,10_50,,,,0.4643,0.2857,0.1071,0.1429,0.464,0.286,0.107,0.143,,,,,,,,,,,
+,0.363,0.091,0.546,0,0,0,8.728,136.815,124.637,1.8646,8528.29,91.921,1551.279,759.0348,1626.9084,3147.317,1.7996,10.92,1,10_50,,,,0.4286,0.2857,0.1429,0.1429,0.429,0.286,0.143,0.143,,,,,,,,,,,
+,0.327,0.091,0.582,0,0,0,8.764,136.635,124.673,1.8664,8565.226,92.641,1572.411,758.9602,1629.9832,3157.2134,1.7816,9.91,1,10_50,,,,0.3929,0.2857,0.1786,0.1429,0.393,0.286,0.179,0.143,,,,,,,,,,,
+,0.273,0.091,0.636,0,0,0,8.818,136.365,124.727,1.8691,8620.63,93.721,1604.109,758.8485,1634.5953,3172.058,1.7546,11.78,1,10_50,,,,0.3571,0.2857,0.2143,0.1429,0.357,0.286,0.214,0.143,,,,,,,,,,,
+,0.227,0.091,0.682,0,0,0,8.864,136.135,124.773,1.8714,8667.826,94.641,1631.111,758.7532,1638.5242,3184.7034,1.7316,12.55,1,10_50,,,,0.3214,0.2857,0.25,0.1429,0.321,0.286,0.25,0.143,,,,,,,,,,,
+,0.355,0.091,0.527,0,0.027,0,8.628,136.775,124.834,1.8582,8461.168,90.2099,1567.18,756.062,1621.2092,3138.6025,1.7491,13.16,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.546,0.091,0.318,0,0.045,0,8.365,137.73,124.769,1.8442,8214.982,85.1425,1462.533,754.4865,1600.6408,3078.8209,1.8137,10.94,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.5,0.091,0.364,0,0.045,0,8.411,137.5,124.815,1.8465,8262.178,86.0625,1489.535,754.3913,1604.5697,3091.4663,1.7907,10.87,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.455,0.091,0.409,0,0.045,0,8.456,137.275,124.86,1.8487,8308.348,86.9625,1515.95,754.2981,1608.4131,3103.8368,1.7682,13.6,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.409,0.091,0.455,0,0.045,0,8.502,137.045,124.906,1.851,8355.544,87.8825,1542.952,754.2029,1612.342,3116.4822,1.7452,14,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.364,0.091,0.5,0,0.045,0,8.547,136.82,124.951,1.8533,8401.714,88.7825,1569.367,754.1098,1616.1855,3128.8527,1.7227,12.19,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.346,0.091,0.518,0,0.045,0,8.565,136.73,124.969,1.8542,8420.182,89.1425,1579.933,754.0725,1617.7228,3133.8009,1.7137,11.92,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.318,0.091,0.546,0,0.045,0,8.593,136.59,124.997,1.8556,8448.91,89.7025,1596.369,754.0146,1620.1143,3141.4981,1.6997,11.49,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.273,0.091,0.591,0,0.045,0,8.638,136.365,125.042,1.8578,8495.08,90.6025,1622.784,753.9214,1623.9578,3153.8686,1.6772,14.07,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.482,0.091,0.363,0,0.064,0,8.353,137.41,124.966,1.8426,8227.636,85.1058,1507.986,752.2737,1601.6157,3088.7345,1.749,14.28,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.454,0.091,0.391,0,0.064,0,8.381,137.27,124.994,1.844,8256.364,85.6658,1524.422,752.2158,1604.0071,3096.4317,1.735,13.97,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.436,0.091,0.409,0,0.064,0,8.399,137.18,125.012,1.8449,8274.832,86.0258,1534.988,752.1785,1605.5445,3101.3799,1.726,10.48,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.409,0.091,0.436,0,0.064,0,8.426,137.045,125.039,1.8463,8302.534,86.5658,1550.837,752.1226,1607.8506,3108.8022,1.7125,4.23,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.391,0.091,0.454,0,0.064,0,8.444,136.955,125.057,1.8472,8321.002,86.9258,1561.403,752.0854,1609.388,3113.7504,1.7035,7.38,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.363,0.091,0.482,0,0.064,0,8.472,136.815,125.085,1.8486,8349.73,87.4858,1577.839,752.0274,1611.7795,3121.4476,1.6895,7.48,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.345,0.091,0.5,0,0.064,0,8.49,136.725,125.103,1.8495,8368.198,87.8458,1588.405,751.9901,1613.3168,3126.3958,1.6805,10.93,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.318,0.091,0.527,0,0.064,0,8.517,136.59,125.13,1.8508,8395.9,88.3858,1604.254,751.9342,1615.6229,3133.8181,1.667,10.85,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.454,0.091,0.382,0,0.073,0,8.345,137.27,125.057,1.8418,8231.254,85.0421,1528.157,751.2304,1601.8796,3092.7938,1.7195,13.88,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.427,0.091,0.409,0,0.073,0,8.372,137.135,125.084,1.8431,8258.956,85.5821,1544.006,751.1745,1604.1857,3100.2161,1.706,4.38,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.409,0.091,0.427,0,0.073,0,8.39,137.045,125.102,1.844,8277.424,85.9421,1554.572,751.1372,1605.7231,3105.1643,1.697,7.59,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.382,0.091,0.454,0,0.073,0,8.417,136.91,125.129,1.8454,8305.126,86.4821,1570.421,751.0813,1608.0292,3112.5866,1.6835,8.59,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.363,0.091,0.473,0,0.073,0,8.436,136.815,125.148,1.8463,8324.62,86.8621,1581.574,751.042,1609.6519,3117.8097,1.674,8.55,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.336,0.091,0.5,0,0.073,0,8.463,136.68,125.175,1.8477,8352.322,87.4021,1597.423,750.9861,1611.958,3125.232,1.6605,10.8,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.291,0.091,0.545,0,0.073,0,8.508,136.455,125.22,1.8499,8398.492,88.3021,1623.838,750.8929,1615.8015,3137.6025,1.638,12.61,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.418,0.091,0.409,0,0.082,0,8.345,137.09,125.156,1.8413,8243.08,85.1384,1553.024,750.1704,1602.8269,3099.0523,1.686,7.72,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.391,0.091,0.436,0,0.082,0,8.372,136.955,125.183,1.8427,8270.782,85.6784,1568.873,750.1145,1605.133,3106.4746,1.6725,6.89,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.372,0.091,0.455,0,0.082,0,8.391,136.86,125.202,1.8436,8290.276,86.0584,1580.026,750.0752,1606.7557,3111.6977,1.663,7.48,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.354,0.091,0.473,0,0.082,0,8.409,136.77,125.22,1.8445,8308.744,86.4184,1590.592,750.0379,1608.2931,3116.6459,1.654,8.5,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.336,0.091,0.491,0,0.082,0,8.427,136.68,125.238,1.8454,8327.212,86.7784,1601.158,750.0007,1609.8305,3121.5941,1.645,9.88,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.318,0.091,0.509,0,0.082,0,8.445,136.59,125.256,1.8463,8345.68,87.1384,1611.724,749.9634,1611.3679,3126.5423,1.636,10.98,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.5,0.091,0.318,0,0.091,0,8.227,137.5,125.137,1.835,8133.838,82.8747,1508.625,749.3548,1593.6958,3072.8726,1.7116,19.79,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.455,0.091,0.363,0,0.091,0,8.272,137.275,125.182,1.8372,8180.008,83.7747,1535.04,749.2616,1597.5392,3085.2431,1.6891,19.27,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.409,0.091,0.409,0,0.091,0,8.318,137.045,125.228,1.8395,8227.204,84.6947,1562.042,749.1664,1601.4681,3097.8885,1.6661,14.3,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.382,0.091,0.436,0,0.091,0,8.345,136.91,125.255,1.8409,8254.906,85.2347,1577.891,749.1105,1603.7741,3105.3108,1.6526,7.81,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.363,0.091,0.455,0,0.091,0,8.364,136.815,125.274,1.8418,8274.4,85.6147,1589.044,749.0712,1605.3969,3110.5339,1.6431,10.27,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.318,0.091,0.5,0,0.091,0,8.409,136.59,125.319,1.8441,8320.57,86.5147,1615.459,748.978,1609.2404,3122.9044,1.6206,10.57,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.273,0.091,0.545,0,0.091,0,8.454,136.365,125.364,1.8463,8366.74,87.4147,1641.874,748.8849,1613.0838,3135.2749,1.5981,13.45,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.345,0.091,0.455,0,0.109,0,8.31,136.725,125.418,1.8382,8242.648,84.7273,1607.08,747.0631,1602.6793,3108.2063,1.6031,17.33,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.464,0.091,0.318,0,0.127,0,8.119,137.32,125.425,1.8278,8070.334,81.0999,1544.697,745.3386,1588.2605,3068.2174,1.6316,20.43,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.418,0.091,0.364,0,0.127,0,8.165,137.09,125.471,1.8301,8117.53,82.0199,1571.699,745.2434,1592.1893,3080.8628,1.6086,19.05,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.373,0.091,0.409,0,0.127,0,8.21,136.865,125.516,1.8323,8163.7,82.9199,1598.114,745.1502,1596.0328,3093.2333,1.5861,20.25,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.318,0.091,0.455,0,0.136,0,8.229,136.59,125.634,1.8328,8195.02,83.3962,1634.134,744.051,1598.6028,3104.7149,1.5432,19.78,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.273,0.091,0.5,0,0.136,0,8.274,136.365,125.679,1.8351,8241.19,84.2962,1660.549,743.9578,1602.4463,3117.0854,1.5207,11.6,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.48,0.2,0.3,0,0.02,0,8.64,137.4,124.66,1.857,8353.32,87.214,1486.54,754.5518,1623.5934,3124.8038,1.7016,13.69,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.455,0.2,0.325,0,0.02,0,8.665,137.275,124.685,1.8583,8378.97,87.714,1501.215,754.5001,1625.7287,3131.6763,1.6891,13.09,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.43,0.2,0.35,0,0.02,0,8.69,137.15,124.71,1.8595,8404.62,88.214,1515.89,754.4483,1627.8639,3138.5488,1.6766,12.11,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.56,0.2,0.2,0,0.04,0,8.48,137.8,124.72,1.848,8215.44,84.228,1447.88,752.5276,1612.0328,3094.7276,1.7072,13.81,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.51,0.2,0.25,0,0.04,0,8.53,137.55,124.77,1.8505,8266.74,85.228,1477.23,752.4241,1616.3033,3108.4726,1.6822,13.58,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.485,0.2,0.275,0,0.04,0,8.555,137.425,124.795,1.8518,8292.39,85.728,1491.905,752.3724,1618.4386,3115.3451,1.6697,6.99,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.46,0.2,0.3,0,0.04,0,8.58,137.3,124.82,1.853,8318.04,86.228,1506.58,752.3206,1620.5738,3122.2176,1.6572,8.28,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.41,0.2,0.35,0,0.04,0,8.63,137.05,124.87,1.8555,8369.34,87.228,1535.93,752.2171,1624.8443,3135.9626,1.6322,11.56,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.36,0.2,0.4,0,0.04,0,8.68,136.8,124.92,1.858,8420.64,88.228,1565.28,752.1136,1629.1148,3149.7076,1.6072,13.08,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.31,0.2,0.45,0,0.04,0,8.73,136.55,124.97,1.8605,8471.94,89.228,1594.63,752.0101,1633.3853,3163.4526,1.5822,13.64,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.25,0.2,0.51,0,0.04,0,8.79,136.25,125.03,1.8635,8533.5,90.428,1629.85,751.8859,1638.5099,3179.9466,1.5522,13.57,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.6,0.2,0.15,0,0.05,0,8.4,138,124.75,1.8435,8146.5,82.735,1428.55,751.5155,1606.2525,3079.6895,1.71,13.61,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.525,0.2,0.225,0,0.05,0,8.475,137.625,124.825,1.8473,8223.45,84.235,1472.575,751.3603,1612.6583,3100.307,1.6725,12.04,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.465,0.2,0.275,0,0.06,0,8.495,137.325,124.955,1.8478,8257.11,84.742,1511.945,750.1412,1615.419,3112.7589,1.6253,5.44,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.415,0.2,0.325,0,0.06,0,8.545,137.075,125.005,1.8503,8308.41,85.742,1541.295,750.0377,1619.6895,3126.5039,1.6003,9.7,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.365,0.2,0.375,0,0.06,0,8.595,136.825,125.055,1.8528,8359.71,86.742,1570.645,749.9342,1623.96,3140.2489,1.5753,11.6,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.315,0.2,0.425,0,0.06,0,8.645,136.575,125.105,1.8553,8411.01,87.742,1599.995,749.8307,1628.2305,3153.9939,1.5503,13.9,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.265,0.2,0.475,0,0.06,0,8.695,136.325,125.155,1.8578,8462.31,88.742,1629.345,749.7272,1632.501,3167.7389,1.5253,14.63,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.58,0.2,0.15,0,0.07,0,8.34,137.9,124.91,1.8395,8111.22,81.749,1448.59,749.2843,1603.2329,3077.1033,1.6656,20.13,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.53,0.2,0.2,0,0.07,0,8.39,137.65,124.96,1.842,8162.52,82.749,1477.94,749.1808,1607.5034,3090.8483,1.6406,7.17,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.505,0.2,0.225,0,0.07,0,8.415,137.525,124.985,1.8433,8188.17,83.249,1492.615,749.1291,1609.6387,3097.7208,1.6281,5.02,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.48,0.2,0.25,0,0.07,0,8.44,137.4,125.01,1.8445,8213.82,83.749,1507.29,749.0773,1611.7739,3104.5933,1.6156,6.24,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.43,0.2,0.3,0,0.07,0,8.49,137.15,125.06,1.847,8265.12,84.749,1536.64,748.9738,1616.0444,3118.3383,1.5906,7.52,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.405,0.2,0.325,0,0.07,0,8.515,137.025,125.085,1.8483,8290.77,85.249,1551.315,748.9221,1618.1797,3125.2108,1.5781,9.15,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.38,0.2,0.35,0,0.07,0,8.54,136.9,125.11,1.8495,8316.42,85.749,1565.99,748.8703,1620.3149,3132.0833,1.5656,10.73,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.355,0.2,0.375,0,0.07,0,8.565,136.775,125.135,1.8508,8342.07,86.249,1580.665,748.8186,1622.4502,3138.9558,1.5531,11.05,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.33,0.2,0.4,0,0.07,0,8.59,136.65,125.16,1.852,8367.72,86.749,1595.34,748.7668,1624.5854,3145.8283,1.5406,11.78,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.28,0.2,0.45,0,0.07,0,8.64,136.4,125.21,1.8545,8419.02,87.749,1624.69,748.6633,1628.8559,3159.5733,1.5156,13.11,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.25,0.2,0.48,0,0.07,0,8.67,136.25,125.24,1.856,8449.8,88.349,1642.3,748.6012,1631.4182,3167.8203,1.5006,14.14,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.445,0.2,0.275,0,0.08,0,8.435,137.225,125.115,1.8438,8221.83,83.756,1531.985,747.91,1612.3994,3110.1727,1.5809,6.74,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.395,0.2,0.325,0,0.08,0,8.485,136.975,125.165,1.8463,8273.13,84.756,1561.335,747.8065,1616.6699,3123.9177,1.5559,8.64,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.37,0.2,0.35,0,0.08,0,8.51,136.85,125.19,1.8475,8298.78,85.256,1576.01,747.7547,1618.8051,3130.7902,1.5434,10.01,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.345,0.2,0.375,0,0.08,0,8.535,136.725,125.215,1.8488,8324.43,85.756,1590.685,747.703,1620.9404,3137.6627,1.5309,11.07,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.295,0.2,0.425,0,0.08,0,8.585,136.475,125.265,1.8513,8375.73,86.756,1620.035,747.5995,1625.2109,3151.4077,1.5059,12.41,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+(?),0.485,0.2,0.225,0,0.09,0,8.355,137.425,125.145,1.8393,8152.89,82.263,1512.655,746.8979,1606.6191,3095.1346,1.5837,6.76,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.41,0.2,0.3,0,0.09,0,8.43,137.05,125.22,1.843,8229.84,83.763,1556.68,746.7426,1613.0248,3115.7521,1.5462,8.33,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.36,0.2,0.35,0,0.09,0,8.48,136.8,125.27,1.8455,8281.14,84.763,1586.03,746.6391,1617.2953,3129.4971,1.5212,10.18,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.31,0.2,0.4,0,0.09,0,8.53,136.55,125.32,1.848,8332.44,85.763,1615.38,746.5356,1621.5658,3143.2421,1.4962,11.79,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.26,0.2,0.45,0,0.09,0,8.58,136.3,125.37,1.8505,8383.74,86.763,1644.73,746.4321,1625.8363,3156.9871,1.4712,13.55,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.55,0.2,0.15,0,0.1,0,8.25,137.75,125.15,1.8335,8058.3,80.27,1478.65,745.9375,1598.7035,3073.224,1.599,20.37,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.5,0.2,0.2,0,0.1,0,8.3,137.5,125.2,1.836,8109.6,81.27,1508,745.834,1602.974,3086.969,1.574,14.32,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.45,0.2,0.25,0,0.1,0,8.35,137.25,125.25,1.8385,8160.9,82.27,1537.35,745.7305,1607.2445,3100.714,1.549,15.76,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.375,0.2,0.325,0,0.1,0,8.425,136.875,125.325,1.8423,8237.85,83.77,1581.375,745.5753,1613.6503,3121.3315,1.5115,10.85,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.325,0.2,0.375,0,0.1,0,8.475,136.625,125.375,1.8448,8289.15,84.77,1610.725,745.4718,1617.9208,3135.0765,1.4865,11.34,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.275,0.2,0.425,0,0.1,0,8.525,136.375,125.425,1.8473,8340.45,85.77,1640.075,745.3683,1622.1913,3148.8215,1.4615,12.94,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.25,0.2,0.45,0,0.1,0,8.55,136.25,125.45,1.8485,8366.1,86.27,1654.75,745.3165,1624.3265,3155.694,1.449,13.2,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.405,0.2,0.275,0,0.12,0,8.315,137.025,125.435,1.8358,8151.27,81.784,1572.065,743.4476,1606.3602,3105.0003,1.4921,14.41,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.38,0.2,0.3,0,0.12,0,8.34,136.9,125.46,1.837,8176.92,82.284,1586.74,743.3958,1608.4954,3111.8728,1.4796,10.09,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.33,0.2,0.35,0,0.12,0,8.39,136.65,125.51,1.8395,8228.22,83.284,1616.09,743.2923,1612.7659,3125.6178,1.4546,11.97,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.47,0.2,0.2,0,0.13,0,8.21,137.35,125.44,1.83,8056.68,79.791,1538.06,742.4872,1598.4446,3083.0897,1.5074,19.61,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.42,0.2,0.25,0,0.13,0,8.26,137.1,125.49,1.8325,8107.98,80.791,1567.41,742.3837,1602.7151,3096.8347,1.4824,19.75,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+(?),0.37,0.2,0.3,0,0.13,0,8.31,136.85,125.54,1.835,8159.28,81.791,1596.76,742.2802,1606.9856,3110.5797,1.4574,18.63,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.32,0.2,0.35,0,0.13,0,8.36,136.6,125.59,1.8375,8210.58,82.791,1626.11,742.1767,1611.2561,3124.3247,1.4324,13.5,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.27,0.2,0.4,0,0.13,0,8.41,136.35,125.64,1.84,8261.88,83.791,1655.46,742.0732,1615.5266,3138.0697,1.4074,14.07,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.5,0.2,0.15,0,0.15,0,8.1,137.5,125.55,1.8235,7970.1,77.805,1528.75,740.3595,1591.1545,3066.7585,1.488,18.41,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.68,0.3,0,0,0.02,0,8.54,138.4,124.46,1.85,8148.92,82.314,1365.14,752.6398,1616.9754,3086.0938,1.6896,17.56,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.655,0.3,0.025,0,0.02,0,8.565,138.275,124.485,1.8513,8174.57,82.814,1379.815,752.5881,1619.1107,3092.9663,1.6771,9.25,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.63,0.3,0.05,0,0.02,0,8.59,138.15,124.51,1.8525,8200.22,83.314,1394.49,752.5363,1621.2459,3099.8388,1.6646,5.11,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.605,0.3,0.075,0,0.02,0,8.615,138.025,124.535,1.8538,8225.87,83.814,1409.165,752.4846,1623.3812,3106.7113,1.6521,1.86,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.58,0.3,0.1,0,0.02,0,8.64,137.9,124.56,1.855,8251.52,84.314,1423.84,752.4328,1625.5164,3113.5838,1.6396,2.57,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.53,0.3,0.15,0,0.02,0,8.69,137.65,124.61,1.8575,8302.82,85.314,1453.19,752.3293,1629.7869,3127.3288,1.6146,7.3,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.505,0.3,0.175,0,0.02,0,8.715,137.525,124.635,1.8588,8328.47,85.814,1467.865,752.2776,1631.9222,3134.2013,1.6021,9.98,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.455,0.3,0.225,0,0.02,0,8.765,137.275,124.685,1.8613,8379.77,86.814,1497.215,752.1741,1636.1927,3147.9463,1.5771,12.38,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.43,0.3,0.25,0,0.02,0,8.79,137.15,124.71,1.8625,8405.42,87.314,1511.89,752.1223,1638.3279,3154.8188,1.5646,13.47,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.38,0.3,0.3,0,0.02,0,8.84,136.9,124.76,1.865,8456.72,88.314,1541.24,752.0188,1642.5984,3168.5638,1.5396,13.94,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.33,0.3,0.35,0,0.02,0,8.89,136.65,124.81,1.8675,8508.02,89.314,1570.59,751.9153,1646.8689,3182.3088,1.5146,15.54,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.635,0.3,0.025,0,0.04,0,8.505,138.175,124.645,1.8473,8139.29,81.828,1399.855,750.3569,1616.0911,3090.3801,1.6327,2.11,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.61,0.3,0.05,0,0.04,0,8.53,138.05,124.67,1.8485,8164.94,82.328,1414.53,750.3051,1618.2263,3097.2526,1.6202,6.44,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.585,0.3,0.075,0,0.04,0,8.555,137.925,124.695,1.8498,8190.59,82.828,1429.205,750.2534,1620.3616,3104.1251,1.6077,5.57,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.56,0.3,0.1,0,0.04,0,8.58,137.8,124.72,1.851,8216.24,83.328,1443.88,750.2016,1622.4968,3110.9976,1.5952,4.36,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.535,0.3,0.125,0,0.04,0,8.605,137.675,124.745,1.8523,8241.89,83.828,1458.555,750.1499,1624.6321,3117.8701,1.5827,6.17,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.485,0.3,0.175,0,0.04,0,8.655,137.425,124.795,1.8548,8293.19,84.828,1487.905,750.0464,1628.9026,3131.6151,1.5577,8.61,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.46,0.3,0.2,0,0.04,0,8.68,137.3,124.82,1.856,8318.84,85.328,1502.58,749.9946,1631.0378,3138.4876,1.5452,12.03,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.435,0.3,0.225,0,0.04,0,8.705,137.175,124.845,1.8573,8344.49,85.828,1517.255,749.9429,1633.1731,3145.3601,1.5327,12.27,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.41,0.3,0.25,0,0.04,0,8.73,137.05,124.87,1.8585,8370.14,86.328,1531.93,749.8911,1635.3083,3152.2326,1.5202,12.01,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.385,0.3,0.275,0,0.04,0,8.755,136.925,124.895,1.8598,8395.79,86.828,1546.605,749.8394,1637.4436,3159.1051,1.5077,12.05,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.36,0.3,0.3,0,0.04,0,8.78,136.8,124.92,1.861,8421.44,87.328,1561.28,749.7876,1639.5788,3165.9776,1.4952,13.26,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.64,0.3,0,0,0.06,0,8.42,138.2,124.78,1.842,8078.36,80.342,1405.22,748.1774,1610.9362,3080.9214,1.6008,23.2,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.59,0.3,0.05,0,0.06,0,8.47,137.95,124.83,1.8445,8129.66,81.342,1434.57,748.0739,1615.2067,3094.6664,1.5758,14.82,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.565,0.3,0.075,0,0.06,0,8.495,137.825,124.855,1.8458,8155.31,81.842,1449.245,748.0222,1617.342,3101.5389,1.5633,9.06,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.515,0.3,0.125,0,0.06,0,8.545,137.575,124.905,1.8483,8206.61,82.842,1478.595,747.9187,1621.6125,3115.2839,1.5383,7.13,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.49,0.3,0.15,0,0.06,0,8.57,137.45,124.93,1.8495,8232.26,83.342,1493.27,747.8669,1623.7477,3122.1564,1.5258,7.5,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.465,0.3,0.175,0,0.06,0,8.595,137.325,124.955,1.8508,8257.91,83.842,1507.945,747.8152,1625.883,3129.0289,1.5133,9.58,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.44,0.3,0.2,0,0.06,0,8.62,137.2,124.98,1.852,8283.56,84.342,1522.62,747.7634,1628.0182,3135.9014,1.5008,9.82,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.39,0.3,0.25,0,0.06,0,8.67,136.95,125.03,1.8545,8334.86,85.342,1551.97,747.6599,1632.2887,3149.6464,1.4758,11.86,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0.3,0.34,0,0.06,0,8.76,136.5,125.12,1.859,8427.2,87.142,1604.8,747.4736,1639.9756,3174.3874,1.4308,15.93,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.595,0.3,0.025,0,0.08,0,8.385,137.975,124.965,1.8393,8068.73,79.856,1439.935,745.8945,1610.0519,3085.2077,1.5439,18.7,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.545,0.3,0.075,0,0.08,0,8.435,137.725,125.015,1.8418,8120.03,80.856,1469.285,745.791,1614.3224,3098.9527,1.5189,16.69,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.495,0.3,0.125,0,0.08,0,8.485,137.475,125.065,1.8443,8171.33,81.856,1498.635,745.6875,1618.5929,3112.6977,1.4939,9.98,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.47,0.3,0.15,0,0.08,0,8.51,137.35,125.09,1.8455,8196.98,82.356,1513.31,745.6357,1620.7281,3119.5702,1.4814,9.75,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.445,0.3,0.175,0,0.08,0,8.535,137.225,125.115,1.8468,8222.63,82.856,1527.985,745.584,1622.8634,3126.4427,1.4689,11.08,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.42,0.3,0.2,0,0.08,0,8.56,137.1,125.14,1.848,8248.28,83.356,1542.66,745.5322,1624.9986,3133.3152,1.4564,9.61,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.395,0.3,0.225,0,0.08,0,8.585,136.975,125.165,1.8493,8273.93,83.856,1557.335,745.4805,1627.1339,3140.1877,1.4439,11.29,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.37,0.3,0.25,0,0.08,0,8.61,136.85,125.19,1.8505,8299.58,84.356,1572.01,745.4287,1629.2691,3147.0602,1.4314,12.02,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.345,0.3,0.275,0,0.08,0,8.635,136.725,125.215,1.8518,8325.23,84.856,1586.685,745.377,1631.4044,3153.9327,1.4189,11.92,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0.3,0.32,0,0.08,0,8.68,136.5,125.26,1.854,8371.4,85.756,1613.1,745.2838,1635.2478,3166.3032,1.3964,15.25,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.6,0.3,0,0,0.1,0,8.3,138,125.1,1.834,8007.8,78.37,1445.3,743.715,1604.897,3075.749,1.512,22.63,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.55,0.3,0.05,0,0.1,0,8.35,137.75,125.15,1.8365,8059.1,79.37,1474.65,743.6115,1609.1675,3089.494,1.487,18.6,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.45,0.3,0.15,0,0.1,0,8.45,137.25,125.25,1.8415,8161.7,81.37,1533.35,743.4045,1617.7085,3116.984,1.437,12.43,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.4,0.3,0.2,0,0.1,0,8.5,137,125.3,1.844,8213,82.37,1562.7,743.301,1621.979,3130.729,1.412,11.63,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.35,0.3,0.25,0,0.1,0,8.55,136.75,125.35,1.8465,8264.3,83.37,1592.05,743.1975,1626.2495,3144.474,1.387,12.28,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.3,0.3,0.3,0,0.1,0,8.6,136.5,125.4,1.849,8315.6,84.37,1621.4,743.094,1630.52,3158.219,1.362,16.84,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.555,0.3,0.025,0,0.12,0,8.265,137.775,125.285,1.8313,7998.17,77.884,1480.015,741.4321,1604.0127,3080.0353,1.4551,20.22,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.505,0.3,0.075,0,0.12,0,8.315,137.525,125.335,1.8338,8049.47,78.884,1509.365,741.3286,1608.2832,3093.7803,1.4301,18.45,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.455,0.3,0.125,0,0.12,0,8.365,137.275,125.385,1.8363,8100.77,79.884,1538.715,741.2251,1612.5537,3107.5253,1.4051,13.25,1,10_50,,,,,,,,,,,,,,,,,,,,,,
+,0.405,0.3,0.175,0,0.12,0,8.415,137.025,125.435,1.8388,8152.07,80.884,1568.065,741.1216,1616.8242,3121.2703,1.3801,16.77,1,10_50,,,0.621,0.286,0.079,0.014,0,0,8.623,138.175,124.379,1.8575,8240.502,84.922,1397.801,753.5277,1624.4696,3104.6865,1.6777,0.54,,
+,0.355,0.3,0.225,0,0.12,0,8.465,136.775,125.485,1.8413,8203.37,81.884,1597.415,741.0181,1621.0947,3135.0153,1.3551,13.22,1,10_50,,,0.428,0.091,0.395,0.086,0,0,8.405,137.57,124.572,1.8613,8310.24,90.105,1548.556,749.9217,1616.3946,3108.3613,1.6326,2.2,1,
+,0.3,0.3,0.28,0,0.12,0,8.52,136.5,125.54,1.844,8259.8,82.984,1629.7,740.9042,1625.7922,3150.1348,1.3276,12.94,1,10_50,,,0.501,0.167,0.261,0.071,0,0,8.453,137.86,124.499,1.86,8262.35,88.051,1496.485,749.918,1618.97777,3104.3368,1.6187,1.69,1,
+,0.23,0.3,0.35,0,0.12,0,8.59,136.15,125.61,1.8475,8331.62,84.384,1670.79,740.7593,1631.7709,3169.3778,1.2926,10.95,1,10_50,,,0.561,0.231,0.162,0.046,0,0,8.532,138.035,124.439,1.8589,8245.302,86.425,1448.405,751.2418,1621.99263,3104.3856,1.6351,1.87,1,
+,0.55,0.3,0,0,0.15,0,8.15,137.75,125.5,1.824,7919.6,75.905,1495.4,738.137,1597.348,3069.2835,1.401,21.03,1,10_50,,Q2,0.5523,0.2386,0.167,0.0421,0,0,8.56,137.972,124.4477,1.8595,8261.153,86.554,1451.6011,751.4664,1623.755991,3108.97003,1.6312,7.6,1,10_50
+,0.5,0.3,0.05,0,0.15,0,8.2,137.5,125.55,1.8265,7970.9,76.905,1524.75,738.0335,1601.6185,3083.0285,1.376,20.7,1,10_50,,Q3,0.4922,0.1724,0.2705,0.0649,0,0,8.4855,137.7855,124.5078,1.8606,8282.158,88.215,1498.9214,750.4301,1620.646404,3109.13022,1.6224,10.39,1,10_50
+FeNiCo_new,0.625,0.34,0.035,0,0,0,8.715,138.125,124.375,1.859,8261.47,84.44,1387.525,753.7854,1630.5864,3115.8055,1.6517,0.3,1,30-100,,Q4,0.4184,0.0937,0.4085,0.0794,0,0,8.4371,137.489,124.5816,1.8619,8331.7272,90.3123,1551.364,750.5487,1617.877844,3113.05795,1.64,0.875,1,10_50
+,0.635,0.325,0.04,0,0,0,8.69,138.175,124.365,1.858,8251.09,84.375,1382.255,754.155,1628.1627,3110.616,1.6735,0.1,1,30-100,,Q5,0.5254,0.2188,0.2083,0.0475,0,0,8.5509,137.8645,124.4746,1.8603,8279.09,87.2378,1470.4082,751.2906,1623.670068,3111.8193,1.6274,8.11,1,10_50
+,0.63,0.33,0.04,0,0,0,8.7,138.15,124.37,1.8584,8256.26,84.43,1384.99,754.0283,1629.1129,3112.804,1.6654,0.4,1,30-100,,Q6,0.3997,0.0686,0.395,0.079,0,0.0577,8.5473,137.3935,124.7734,1.8632,8352.5706,108.2246,1539.5231,750.2933,1634.731508,3132.8096,1.5605,6.04,1,10_50
+,0.625,0.335,0.04,0,0,0,8.71,138.125,124.375,1.8588,8261.43,84.485,1387.725,753.9017,1630.0632,3114.992,1.6573,0.5,1,30-100,,Q7,0.4404,0.1371,0.3467,0.0758,0,0,8.4693,137.581,124.5596,1.8621,8315.8384,89.5033,1535.2308,749.9719,1620.74792,3114.95405,1.6108,3.34,1,10_50
+,0.635,0.315,0.05,0,0,0,8.68,138.175,124.365,1.8577,8251.01,84.465,1382.655,754.3876,1627.1163,3108.989,1.6847,0.1,1,30-100,,Q8,0.562,0.231,0.161,0.046,0,0,8.531,138.04,124.438,1.8588,8244.276,86.405,1447.818,751.2439,1621.9072,3104.1107,1.6356,8.38,1,10_50
+,0.625,0.325,0.05,0,0,0,8.7,138.125,124.375,1.8585,8261.35,84.575,1388.125,754.1343,1629.0168,3113.365,1.6685,0.5,1,30-100,,Q9,0.501,0.167,0.261,0.071,0,0,8.453,137.86,124.499,1.86,8262.35,88.051,1496.485,749.91802,1618.9778,3104.3368,1.6187,1.75,1,10_50
+,0.635,0.305,0.06,0,0,0,8.67,138.175,124.365,1.8574,8250.93,84.555,1383.055,754.6202,1626.0699,3107.362,1.6959,0.1,1,30-100,,Q10,0.427,0.091,0.396,0.086,0,0,8.406,137.565,124.573,1.8614,8311.266,90.125,1549.143,749.91965,1616.48,3108.6362,1.6321,1.32,1,10_50
+,0.625,0.315,0.06,0,0,0,8.69,138.125,124.375,1.8582,8261.27,84.665,1388.525,754.3669,1627.9704,3111.738,1.6797,0.1,1,30-100,,Q11,0.602,0.2407,0.0927,0.0075,0,0.0568,8.7274,138.0079,124.532,1.8577,8269.6701,102.7722,1384.2527,754.1914773,1638.8419,3121.517272,1.6358,8.78,0.9997,10_50
diff --git a/Tutorial_NN.ipynb b/Tutorial_NN.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..e4f2d22676437eb88397788cca95cb9e562a0ad0
--- /dev/null
+++ b/Tutorial_NN.ipynb
@@ -0,0 +1,3677 @@
+{
+  "nbformat": 4,
+  "nbformat_minor": 0,
+  "metadata": {
+    "colab": {
+      "name": "Tutorial_BigMax.ipynb",
+      "provenance": [],
+      "collapsed_sections": [],
+      "toc_visible": true
+    },
+    "kernelspec": {
+      "name": "python3",
+      "display_name": "Python 3"
+    },
+    "language_info": {
+      "name": "python"
+    }
+  },
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "aOpB14foyoVj"
+      },
+      "source": [
+        "#Introduction\n",
+        "In this notebook, we will implement a machine-learning framework accelerates functional high-entropy alloy discovery"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "HO3WeAGHmW_t"
+      },
+      "source": [
+        "import cv2\n",
+        "import os\n",
+        "import time\n",
+        "import random\n",
+        "import numpy as np\n",
+        "import pandas as pd\n",
+        "\n",
+        "import torch\n",
+        "import torch.nn as nn\n",
+        "import torch.nn.functional as F\n",
+        "from torch.optim import Adam, lr_scheduler\n",
+        "from torch.utils.data import Dataset, DataLoader\n",
+        "from scipy.spatial.distance import cdist\n",
+        "from sklearn.model_selection import train_test_split\n",
+        "from sklearn.model_selection import KFold\n",
+        "from sklearn.cluster import KMeans\n",
+        "from sklearn.mixture import GaussianMixture\n",
+        "\n",
+        "import matplotlib.pyplot as plt\n",
+        "import matplotlib.patches as patches\n",
+        "from matplotlib.patches import Ellipse\n",
+        "from mpl_toolkits.axes_grid1 import make_axes_locatable\n",
+        "import seaborn as sns\n",
+        "\n",
+        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
+        "root = '/content/'\n",
+        "\n",
+        "sns.set(color_codes=True)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "OQnO2Nrq9Bc-"
+      },
+      "source": [
+        "#Required packages"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "dTzbyKG1o_0i",
+        "outputId": "53cc21f5-4fa5-4697-9d7d-524ff42e52d8"
+      },
+      "source": [
+        "!pip install bayesian-optimization\n",
+        "!pip install lightgbm"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Collecting bayesian-optimization\n",
+            "  Downloading bayesian-optimization-1.2.0.tar.gz (14 kB)\n",
+            "Requirement already satisfied: numpy>=1.9.0 in /usr/local/lib/python3.7/dist-packages (from bayesian-optimization) (1.19.5)\n",
+            "Requirement already satisfied: scipy>=0.14.0 in /usr/local/lib/python3.7/dist-packages (from bayesian-optimization) (1.4.1)\n",
+            "Requirement already satisfied: scikit-learn>=0.18.0 in /usr/local/lib/python3.7/dist-packages (from bayesian-optimization) (0.22.2.post1)\n",
+            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn>=0.18.0->bayesian-optimization) (1.0.1)\n",
+            "Building wheels for collected packages: bayesian-optimization\n",
+            "  Building wheel for bayesian-optimization (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
+            "  Created wheel for bayesian-optimization: filename=bayesian_optimization-1.2.0-py3-none-any.whl size=11685 sha256=2df350d24e084c203ca5cde681b1b4e4d917800b250f8d76d90c5251bc8962bd\n",
+            "  Stored in directory: /root/.cache/pip/wheels/fd/9b/71/f127d694e02eb40bcf18c7ae9613b88a6be4470f57a8528c5b\n",
+            "Successfully built bayesian-optimization\n",
+            "Installing collected packages: bayesian-optimization\n",
+            "Successfully installed bayesian-optimization-1.2.0\n",
+            "Requirement already satisfied: lightgbm in /usr/local/lib/python3.7/dist-packages (2.2.3)\n",
+            "Requirement already satisfied: scikit-learn in /usr/local/lib/python3.7/dist-packages (from lightgbm) (0.22.2.post1)\n",
+            "Requirement already satisfied: numpy in /usr/local/lib/python3.7/dist-packages (from lightgbm) (1.19.5)\n",
+            "Requirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from lightgbm) (1.4.1)\n",
+            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.7/dist-packages (from scikit-learn->lightgbm) (1.0.1)\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "4lKoaSz8pTHk"
+      },
+      "source": [
+        "#Wasserstein Autoencoder\n",
+        "Wasserstein autoencoder is used for learning the latent space representation"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "hlnrI--Mc9-T"
+      },
+      "source": [
+        "class FeatureDataset(Dataset): #from numpy to tensor (pytroch-readable)\n",
+        "    '''\n",
+        "    Args: x is a 2D numpy array [x_size, x_features]\n",
+        "    '''\n",
+        "    def __init__(self, x, y):\n",
+        "        self.x = x\n",
+        "        self.y = y\n",
+        "    \n",
+        "    def __len__(self):\n",
+        "        return self.x.shape[0]\n",
+        "    \n",
+        "    def __getitem__(self, idx):\n",
+        "        return torch.FloatTensor(self.x[idx]), torch.FloatTensor(self.y[idx])\n",
+        "\n",
+        "class AttributeDataset(Dataset): # this is for classifier \n",
+        "    '''\n",
+        "    Args: x is a 2D numpy array [x_size, x_features]\n",
+        "    '''\n",
+        "    def __init__(self, x, y):\n",
+        "        self.x = x\n",
+        "        self.y = y\n",
+        "    def __len__(self):\n",
+        "        return self.x.shape[0]\n",
+        "    def __getitem__(self, idx):\n",
+        "        return torch.Tensor(self.x[idx]), torch.Tensor(self.y[idx])"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "3OkOhFJf1MBC"
+      },
+      "source": [
+        "**Model architecture**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "IYHg-od4o9dZ"
+      },
+      "source": [
+        "def weights_init(m):\n",
+        "    classname = m.__class__.__name__\n",
+        "    if classname.find('BatchNorm') != -1:\n",
+        "        m.weight.data.normal_(1.0, 0.02)\n",
+        "        m.bias.data.fill_(0)\n",
+        "\n",
+        "class WAE(nn.Module):\n",
+        "    def __init__(self, input_size):\n",
+        "        super(WAE, self).__init__()\n",
+        "        self.input_size = input_size\n",
+        "\n",
+        "        # encoder\n",
+        "        self.encoder = nn.Sequential(\n",
+        "                        nn.Linear(self.input_size, 80),\n",
+        "                        nn.LayerNorm(80),\n",
+        "                        nn.ReLU(),\n",
+        "                        nn.Linear(80, 64),\n",
+        "                        nn.LayerNorm(64),\n",
+        "                        nn.ReLU(),\n",
+        "                        nn.Linear(64, 48),\n",
+        "                        nn.LayerNorm(48),\n",
+        "                        nn.ReLU(),\n",
+        "                        nn.Linear(48, 2),\n",
+        "                        )\n",
+        "\n",
+        "        # decoder\n",
+        "        self.decoder = nn.Sequential(\n",
+        "                        nn.Linear(2, 48),\n",
+        "                        nn.LayerNorm(48),\n",
+        "                        nn.ReLU(),\n",
+        "                        nn.Linear(48, 64),\n",
+        "                        nn.LayerNorm(64),\n",
+        "                        nn.ReLU(),\n",
+        "                        nn.Linear(64, 80),\n",
+        "                        nn.LayerNorm(80),\n",
+        "                        nn.ReLU(),\n",
+        "                        nn.Linear(80, self.input_size),\n",
+        "                        nn.Softmax(dim=1) #(softmad along dimension 1)\n",
+        "                        )\n",
+        "        self.apply(weights_init)\n",
+        "    \n",
+        "    def forward(self, x):\n",
+        "        z = self._encode(x)\n",
+        "        x_recon = self._decode(z)\n",
+        "\n",
+        "        return x_recon, z\n",
+        "    \n",
+        "    def _encode(self, x):\n",
+        "        return self.encoder(x)\n",
+        "\n",
+        "    def _decode(self, z):\n",
+        "        return self.decoder(z)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "Rfgq7_H51Sgi"
+      },
+      "source": [
+        "**Utility functions for WAE**\n",
+        "\n",
+        "\n",
+        "*   same_seeds: fixing the randomness\n",
+        "*   get_latents: get the latent spaces from the WAE\n",
+        "*   imq_kernels: inverse multiquadric (IMQ) kernel - computing the maximum mean discrepancy, which is part of the loss function for WAE.\n",
+        "\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "8yBgwL4mqPFk"
+      },
+      "source": [
+        "def same_seeds(seed): #fix np & torch seed to the same.\n",
+        "    torch.manual_seed(seed)\n",
+        "    if torch.cuda.is_available():\n",
+        "        torch.cuda.manual_seed(seed)\n",
+        "        torch.cuda.manual_seed_all(seed)  # if you are using multi-GPU.\n",
+        "    np.random.seed(seed)  # Numpy module.\n",
+        "    random.seed(seed)  # Python random module.\n",
+        "    torch.backends.cudnn.benchmark = False\n",
+        "    torch.backends.cudnn.deterministic = True\n",
+        "\n",
+        "def get_latents(model, dataset): #from dataset to altten\n",
+        "    model.to(device).eval() # training model or evaluation mode, eval means setting the model to its evaluation mode (gradient fixed)\n",
+        "    latents = []\n",
+        "    with torch.no_grad(): # fix the gradient, assure that the model parameters are fixed\n",
+        "        dataloader = DataLoader(dataset, batch_size=256, shuffle=False)\n",
+        "        for i, data in enumerate(dataloader):\n",
+        "            x = data[0].to(device)\n",
+        "            recon_x, z = model(x)\n",
+        "            latents.append(z.detach().cpu().numpy())\n",
+        "    return np.concatenate(latents,axis=0)\n",
+        "\n",
+        "def imq_kernel(X: torch.Tensor, Y: torch.Tensor, h_dim: int): # common kerntl to choose\n",
+        "    batch_size = X.size(0)\n",
+        "\n",
+        "    norms_x = X.pow(2).sum(1, keepdim=True)  # batch_size x 1\n",
+        "    prods_x = torch.mm(X, X.t()).to(device)  # batch_size x batch_size\n",
+        "    dists_x = norms_x + norms_x.t() - 2 * prods_x # mm matrix multiplicaiton\n",
+        "\n",
+        "    norms_y = Y.pow(2).sum(1, keepdim=True).to(device)  # batch_size x 1\n",
+        "    prods_y = torch.mm(Y, Y.t()).to(device)  # batch_size x batch_size\n",
+        "    dists_y = norms_y + norms_y.t() - 2 * prods_y\n",
+        "\n",
+        "    dot_prd = torch.mm(X, Y.t())\n",
+        "    dists_c = norms_x + norms_y.t() - 2 * dot_prd\n",
+        "\n",
+        "    stats = 0\n",
+        "    for scale in [.1, .2, .5, 1., 2., 5., 10.]: # need more study on this\n",
+        "        C = 2 * h_dim * 1.0 * scale\n",
+        "        res1 = C / (C + dists_x)\n",
+        "        res1 += C / (C + dists_y)\n",
+        "\n",
+        "        if torch.cuda.is_available():\n",
+        "            res1 = (1 - torch.eye(batch_size).to(device)) * res1\n",
+        "        else:\n",
+        "            res1 = (1 - torch.eye(batch_size)) * res1\n",
+        "\n",
+        "        res1 = res1.sum() / (batch_size - 1)\n",
+        "        res2 = C / (C + dists_c)\n",
+        "        res2 = res2.sum() * 2. / (batch_size)\n",
+        "        stats += res1 - res2\n",
+        "\n",
+        "    return stats\n"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "Tf5oEkuZ4fmv"
+      },
+      "source": [
+        "**Data loading**  "
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "KReugJAwrE14",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "62641da2-34dc-49b6-f890-4a31f9228a10"
+      },
+      "source": [
+        "same_seeds(1) #seed equals to 2\n",
+        "\n",
+        "params = {\n",
+        "    'num_epoch' : 200,\n",
+        "    'batch_size' : 20,\n",
+        "    'lr' : 5e-4,\n",
+        "    'weight_decay' : 0.0,\n",
+        "    'sigma' : 8.0,\n",
+        "    'MMD_lambda' : 1e-4,\n",
+        "    'model_name' : 'WAE_v1',\n",
+        "} # for WAE training\n",
+        "all = pd.read_csv('data_base.csv', header=0).iloc[:,1:19].to_numpy()\n",
+        "raw_x = all[:696,:6]\n",
+        "raw_y = all[:696,17].reshape(-1,1)\n",
+        "dataset = FeatureDataset(raw_x[:], raw_y[:]) #numpy to tensor\n",
+        "dataloader = DataLoader(dataset, batch_size=params['batch_size'], shuffle=True) # tensor to dataloader\n",
+        "print(raw_x[50:55])"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "[[0.701 0.2   0.099 0.    0.    0.   ]\n",
+            " [0.65  0.25  0.1   0.    0.    0.   ]\n",
+            " [0.62  0.28  0.1   0.    0.    0.   ]\n",
+            " [0.601 0.3   0.099 0.    0.    0.   ]\n",
+            " [0.58  0.32  0.1   0.    0.    0.   ]]\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "AkcWFeljrTEi",
+        "outputId": "de8eca67-1aea-4ff5-b67e-d7da8abba178"
+      },
+      "source": [
+        "model = WAE(raw_x.shape[1]).to(device) # initialize the model \n",
+        "optimizer = Adam(model.parameters(), lr = params['lr'], weight_decay = params['weight_decay']) # optimizer\n",
+        "def train_WAE(model, optimizer, dataloader, params):\n",
+        "    model_name = params['model_name']\n",
+        "    num_epoch = params['num_epoch']\n",
+        "    sigma = params['sigma'] # assuming the latent space follows Gaussian\n",
+        "    MMD_lambda = params['MMD_lambda'] #WAE distance (maximum mean discrepancy)\n",
+        "\n",
+        "    folder_dir = os.path.join(root, model_name) # a folder to save models\n",
+        "    if not os.path.isdir(folder_dir):\n",
+        "        os.mkdir(folder_dir)\n",
+        "\n",
+        "    for epoch in range(num_epoch):\n",
+        "        start_time = time.time()\n",
+        "        total_loss = [] #save for plot, recon loss+MMD\n",
+        "        total_recon = [] # binary cross entropy\n",
+        "        total_MMD = [] #maximum mean discrepancy\n",
+        "        \n",
+        "        for i, data in enumerate(dataloader):\n",
+        "            x = data[0].to(device)\n",
+        "            y = data[1].to(device)\n",
+        "            model.train() # model goes to train mode\n",
+        "            recon_x, z_tilde = model(x) # latent space is Z_tilde\n",
+        "            z = sigma*torch.randn(z_tilde.size()).to(device) # z is sampled from a Gaussian that has the same dimension (but no relation to z_tilde).\n",
+        "\n",
+        "            recon_loss = F.binary_cross_entropy(recon_x, x, reduction='mean') #lowest reconstruction loss \n",
+        "            #recon_loss = F.mse_loss(recon_x, x, reduction='mean') \n",
+        "            #recon_loss = F.l1_loss(recon_x, x, reduction='mean')\n",
+        "            \n",
+        "            MMD_loss = imq_kernel(z_tilde, z, h_dim=2).to(device) #W-distance between z_tilde and z\n",
+        "            MMD_loss = MMD_loss / x.size(0) #averaging, because recon loss is mean.\n",
+        "            loss = recon_loss + MMD_loss * MMD_lambda #MM_lambda: learning-rate alike, hyperparamer\n",
+        "\n",
+        "            optimizer.zero_grad()\n",
+        "            loss.backward()\n",
+        "            optimizer.step()\n",
+        "\n",
+        "            total_loss.append(loss.item())# from tensor to values\n",
+        "            total_recon.append(recon_loss.item())\n",
+        "            total_MMD.append(MMD_loss.item())\n",
+        "\n",
+        "        avg_loss = sum(total_loss)/len(total_loss)\n",
+        "        avg_recon = sum(total_recon)/len(total_recon)\n",
+        "        avg_MMD = sum(total_MMD)/len(total_MMD)\n",
+        "\n",
+        "        #scheduler.step(avg_loss)\n",
+        "\n",
+        "        print('[{:03}/{:03}] loss: {:.6f} Recon_loss: {:.6f}, MMD_loss:{:.6f}, time: {:.3f} sec'.format(\\\n",
+        "                                        epoch+1, num_epoch, \\\n",
+        "                                        avg_loss, \\\n",
+        "                                        avg_recon, avg_MMD, time.time() - start_time))\n",
+        "        # save the model every 2 epoches\n",
+        "        if (epoch+1) % 1 == 0:\n",
+        "            save_model_dir = str(model_name + \"_{}.pth\".format(epoch+1))\n",
+        "            torch.save(model.state_dict(), os.path.join(folder_dir, save_model_dir))\n",
+        "\n",
+        "train_WAE(model, optimizer, dataloader, params)"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "[001/200] loss: 0.365267 Recon_loss: 0.364768, MMD_loss:4.989809, time: 0.511 sec\n",
+            "[002/200] loss: 0.319626 Recon_loss: 0.319206, MMD_loss:4.200720, time: 0.254 sec\n",
+            "[003/200] loss: 0.311641 Recon_loss: 0.311190, MMD_loss:4.518825, time: 0.254 sec\n",
+            "[004/200] loss: 0.308817 Recon_loss: 0.308352, MMD_loss:4.641864, time: 0.274 sec\n",
+            "[005/200] loss: 0.307272 Recon_loss: 0.306801, MMD_loss:4.709693, time: 0.282 sec\n",
+            "[006/200] loss: 0.306058 Recon_loss: 0.305591, MMD_loss:4.674590, time: 0.289 sec\n",
+            "[007/200] loss: 0.305077 Recon_loss: 0.304605, MMD_loss:4.715405, time: 0.264 sec\n",
+            "[008/200] loss: 0.304452 Recon_loss: 0.303981, MMD_loss:4.709535, time: 0.260 sec\n",
+            "[009/200] loss: 0.304019 Recon_loss: 0.303537, MMD_loss:4.816010, time: 0.260 sec\n",
+            "[010/200] loss: 0.302840 Recon_loss: 0.302368, MMD_loss:4.726309, time: 0.278 sec\n",
+            "[011/200] loss: 0.302216 Recon_loss: 0.301760, MMD_loss:4.558619, time: 0.275 sec\n",
+            "[012/200] loss: 0.301502 Recon_loss: 0.301036, MMD_loss:4.659072, time: 0.254 sec\n",
+            "[013/200] loss: 0.300504 Recon_loss: 0.300046, MMD_loss:4.573521, time: 0.251 sec\n",
+            "[014/200] loss: 0.300207 Recon_loss: 0.299749, MMD_loss:4.582523, time: 0.248 sec\n",
+            "[015/200] loss: 0.300509 Recon_loss: 0.300049, MMD_loss:4.600637, time: 0.253 sec\n",
+            "[016/200] loss: 0.299701 Recon_loss: 0.299238, MMD_loss:4.636392, time: 0.261 sec\n",
+            "[017/200] loss: 0.299085 Recon_loss: 0.298628, MMD_loss:4.566309, time: 0.253 sec\n",
+            "[018/200] loss: 0.299626 Recon_loss: 0.299173, MMD_loss:4.533172, time: 0.277 sec\n",
+            "[019/200] loss: 0.299297 Recon_loss: 0.298853, MMD_loss:4.437655, time: 0.282 sec\n",
+            "[020/200] loss: 0.298892 Recon_loss: 0.298443, MMD_loss:4.487563, time: 0.263 sec\n",
+            "[021/200] loss: 0.298809 Recon_loss: 0.298368, MMD_loss:4.406757, time: 0.263 sec\n",
+            "[022/200] loss: 0.298677 Recon_loss: 0.298238, MMD_loss:4.394921, time: 0.281 sec\n",
+            "[023/200] loss: 0.298359 Recon_loss: 0.297928, MMD_loss:4.307141, time: 0.287 sec\n",
+            "[024/200] loss: 0.298143 Recon_loss: 0.297704, MMD_loss:4.392592, time: 0.258 sec\n",
+            "[025/200] loss: 0.298241 Recon_loss: 0.297804, MMD_loss:4.370174, time: 0.269 sec\n",
+            "[026/200] loss: 0.298149 Recon_loss: 0.297727, MMD_loss:4.223033, time: 0.258 sec\n",
+            "[027/200] loss: 0.297862 Recon_loss: 0.297431, MMD_loss:4.308488, time: 0.255 sec\n",
+            "[028/200] loss: 0.298102 Recon_loss: 0.297673, MMD_loss:4.293345, time: 0.261 sec\n",
+            "[029/200] loss: 0.298092 Recon_loss: 0.297660, MMD_loss:4.323399, time: 0.278 sec\n",
+            "[030/200] loss: 0.297730 Recon_loss: 0.297304, MMD_loss:4.252688, time: 0.260 sec\n",
+            "[031/200] loss: 0.297715 Recon_loss: 0.297287, MMD_loss:4.283416, time: 0.283 sec\n",
+            "[032/200] loss: 0.298159 Recon_loss: 0.297734, MMD_loss:4.251704, time: 0.269 sec\n",
+            "[033/200] loss: 0.297438 Recon_loss: 0.297019, MMD_loss:4.196932, time: 0.267 sec\n",
+            "[034/200] loss: 0.297304 Recon_loss: 0.296881, MMD_loss:4.223873, time: 0.251 sec\n",
+            "[035/200] loss: 0.297258 Recon_loss: 0.296839, MMD_loss:4.193834, time: 0.239 sec\n",
+            "[036/200] loss: 0.297260 Recon_loss: 0.296836, MMD_loss:4.247362, time: 0.254 sec\n",
+            "[037/200] loss: 0.297012 Recon_loss: 0.296596, MMD_loss:4.159245, time: 0.240 sec\n",
+            "[038/200] loss: 0.297512 Recon_loss: 0.297080, MMD_loss:4.314337, time: 0.238 sec\n",
+            "[039/200] loss: 0.297447 Recon_loss: 0.297043, MMD_loss:4.043144, time: 0.234 sec\n",
+            "[040/200] loss: 0.297090 Recon_loss: 0.296682, MMD_loss:4.082294, time: 0.262 sec\n",
+            "[041/200] loss: 0.297027 Recon_loss: 0.296624, MMD_loss:4.034122, time: 0.263 sec\n",
+            "[042/200] loss: 0.297027 Recon_loss: 0.296616, MMD_loss:4.110546, time: 0.259 sec\n",
+            "[043/200] loss: 0.296916 Recon_loss: 0.296500, MMD_loss:4.159121, time: 0.236 sec\n",
+            "[044/200] loss: 0.297510 Recon_loss: 0.297098, MMD_loss:4.128248, time: 0.232 sec\n",
+            "[045/200] loss: 0.297697 Recon_loss: 0.297284, MMD_loss:4.128303, time: 0.256 sec\n",
+            "[046/200] loss: 0.296852 Recon_loss: 0.296450, MMD_loss:4.020794, time: 0.234 sec\n",
+            "[047/200] loss: 0.296569 Recon_loss: 0.296175, MMD_loss:3.939298, time: 0.250 sec\n",
+            "[048/200] loss: 0.296947 Recon_loss: 0.296546, MMD_loss:4.011335, time: 0.275 sec\n",
+            "[049/200] loss: 0.297038 Recon_loss: 0.296639, MMD_loss:3.991586, time: 0.274 sec\n",
+            "[050/200] loss: 0.296728 Recon_loss: 0.296342, MMD_loss:3.851469, time: 0.278 sec\n",
+            "[051/200] loss: 0.296592 Recon_loss: 0.296176, MMD_loss:4.164172, time: 0.259 sec\n",
+            "[052/200] loss: 0.296654 Recon_loss: 0.296247, MMD_loss:4.069011, time: 0.256 sec\n",
+            "[053/200] loss: 0.296471 Recon_loss: 0.296070, MMD_loss:4.012771, time: 0.255 sec\n",
+            "[054/200] loss: 0.296509 Recon_loss: 0.296127, MMD_loss:3.820905, time: 0.242 sec\n",
+            "[055/200] loss: 0.296890 Recon_loss: 0.296500, MMD_loss:3.893738, time: 0.272 sec\n",
+            "[056/200] loss: 0.296556 Recon_loss: 0.296162, MMD_loss:3.936125, time: 0.249 sec\n",
+            "[057/200] loss: 0.296305 Recon_loss: 0.295903, MMD_loss:4.019066, time: 0.275 sec\n",
+            "[058/200] loss: 0.296596 Recon_loss: 0.296194, MMD_loss:4.020099, time: 0.257 sec\n",
+            "[059/200] loss: 0.296607 Recon_loss: 0.296209, MMD_loss:3.976587, time: 0.247 sec\n",
+            "[060/200] loss: 0.296604 Recon_loss: 0.296209, MMD_loss:3.944052, time: 0.266 sec\n",
+            "[061/200] loss: 0.296639 Recon_loss: 0.296240, MMD_loss:3.993885, time: 0.261 sec\n",
+            "[062/200] loss: 0.296476 Recon_loss: 0.296076, MMD_loss:3.993161, time: 0.248 sec\n",
+            "[063/200] loss: 0.296565 Recon_loss: 0.296176, MMD_loss:3.895723, time: 0.259 sec\n",
+            "[064/200] loss: 0.296641 Recon_loss: 0.296251, MMD_loss:3.898492, time: 0.256 sec\n",
+            "[065/200] loss: 0.296226 Recon_loss: 0.295838, MMD_loss:3.881132, time: 0.246 sec\n",
+            "[066/200] loss: 0.296334 Recon_loss: 0.295945, MMD_loss:3.884095, time: 0.273 sec\n",
+            "[067/200] loss: 0.296361 Recon_loss: 0.295970, MMD_loss:3.911597, time: 0.241 sec\n",
+            "[068/200] loss: 0.296077 Recon_loss: 0.295709, MMD_loss:3.680451, time: 0.266 sec\n",
+            "[069/200] loss: 0.296236 Recon_loss: 0.295855, MMD_loss:3.811443, time: 0.231 sec\n",
+            "[070/200] loss: 0.296176 Recon_loss: 0.295794, MMD_loss:3.827666, time: 0.266 sec\n",
+            "[071/200] loss: 0.296770 Recon_loss: 0.296387, MMD_loss:3.837126, time: 0.264 sec\n",
+            "[072/200] loss: 0.296268 Recon_loss: 0.295879, MMD_loss:3.891558, time: 0.243 sec\n",
+            "[073/200] loss: 0.296253 Recon_loss: 0.295877, MMD_loss:3.756104, time: 0.228 sec\n",
+            "[074/200] loss: 0.296685 Recon_loss: 0.296298, MMD_loss:3.872647, time: 0.231 sec\n",
+            "[075/200] loss: 0.296411 Recon_loss: 0.296024, MMD_loss:3.864882, time: 0.238 sec\n",
+            "[076/200] loss: 0.296326 Recon_loss: 0.295938, MMD_loss:3.875427, time: 0.235 sec\n",
+            "[077/200] loss: 0.296403 Recon_loss: 0.296025, MMD_loss:3.779450, time: 0.236 sec\n",
+            "[078/200] loss: 0.296367 Recon_loss: 0.295990, MMD_loss:3.766702, time: 0.234 sec\n",
+            "[079/200] loss: 0.296020 Recon_loss: 0.295643, MMD_loss:3.774297, time: 0.240 sec\n",
+            "[080/200] loss: 0.295899 Recon_loss: 0.295515, MMD_loss:3.835998, time: 0.263 sec\n",
+            "[081/200] loss: 0.296099 Recon_loss: 0.295722, MMD_loss:3.771091, time: 0.245 sec\n",
+            "[082/200] loss: 0.296257 Recon_loss: 0.295884, MMD_loss:3.735502, time: 0.237 sec\n",
+            "[083/200] loss: 0.296104 Recon_loss: 0.295730, MMD_loss:3.744455, time: 0.237 sec\n",
+            "[084/200] loss: 0.296117 Recon_loss: 0.295741, MMD_loss:3.764867, time: 0.237 sec\n",
+            "[085/200] loss: 0.296222 Recon_loss: 0.295840, MMD_loss:3.817816, time: 0.237 sec\n",
+            "[086/200] loss: 0.296218 Recon_loss: 0.295836, MMD_loss:3.816018, time: 0.251 sec\n",
+            "[087/200] loss: 0.296060 Recon_loss: 0.295685, MMD_loss:3.749452, time: 0.240 sec\n",
+            "[088/200] loss: 0.295801 Recon_loss: 0.295425, MMD_loss:3.754975, time: 0.235 sec\n",
+            "[089/200] loss: 0.295898 Recon_loss: 0.295531, MMD_loss:3.666029, time: 0.262 sec\n",
+            "[090/200] loss: 0.295979 Recon_loss: 0.295602, MMD_loss:3.775433, time: 0.239 sec\n",
+            "[091/200] loss: 0.295765 Recon_loss: 0.295393, MMD_loss:3.722068, time: 0.234 sec\n",
+            "[092/200] loss: 0.295782 Recon_loss: 0.295410, MMD_loss:3.725398, time: 0.228 sec\n",
+            "[093/200] loss: 0.296124 Recon_loss: 0.295753, MMD_loss:3.716655, time: 0.279 sec\n",
+            "[094/200] loss: 0.296151 Recon_loss: 0.295781, MMD_loss:3.699157, time: 0.232 sec\n",
+            "[095/200] loss: 0.295971 Recon_loss: 0.295599, MMD_loss:3.718835, time: 0.241 sec\n",
+            "[096/200] loss: 0.295873 Recon_loss: 0.295499, MMD_loss:3.739591, time: 0.236 sec\n",
+            "[097/200] loss: 0.295614 Recon_loss: 0.295245, MMD_loss:3.684349, time: 0.248 sec\n",
+            "[098/200] loss: 0.295839 Recon_loss: 0.295472, MMD_loss:3.673418, time: 0.253 sec\n",
+            "[099/200] loss: 0.295825 Recon_loss: 0.295452, MMD_loss:3.728629, time: 0.261 sec\n",
+            "[100/200] loss: 0.295856 Recon_loss: 0.295477, MMD_loss:3.791965, time: 0.248 sec\n",
+            "[101/200] loss: 0.295877 Recon_loss: 0.295510, MMD_loss:3.666985, time: 0.261 sec\n",
+            "[102/200] loss: 0.295792 Recon_loss: 0.295427, MMD_loss:3.654937, time: 0.254 sec\n",
+            "[103/200] loss: 0.295747 Recon_loss: 0.295371, MMD_loss:3.761321, time: 0.247 sec\n",
+            "[104/200] loss: 0.295733 Recon_loss: 0.295370, MMD_loss:3.630066, time: 0.249 sec\n",
+            "[105/200] loss: 0.295842 Recon_loss: 0.295476, MMD_loss:3.657971, time: 0.254 sec\n",
+            "[106/200] loss: 0.295485 Recon_loss: 0.295122, MMD_loss:3.638396, time: 0.255 sec\n",
+            "[107/200] loss: 0.295845 Recon_loss: 0.295495, MMD_loss:3.493962, time: 0.256 sec\n",
+            "[108/200] loss: 0.295929 Recon_loss: 0.295566, MMD_loss:3.632295, time: 0.259 sec\n",
+            "[109/200] loss: 0.295915 Recon_loss: 0.295546, MMD_loss:3.691325, time: 0.250 sec\n",
+            "[110/200] loss: 0.295672 Recon_loss: 0.295309, MMD_loss:3.635942, time: 0.243 sec\n",
+            "[111/200] loss: 0.295870 Recon_loss: 0.295512, MMD_loss:3.578719, time: 0.247 sec\n",
+            "[112/200] loss: 0.295908 Recon_loss: 0.295556, MMD_loss:3.519477, time: 0.231 sec\n",
+            "[113/200] loss: 0.295623 Recon_loss: 0.295264, MMD_loss:3.581447, time: 0.249 sec\n",
+            "[114/200] loss: 0.295689 Recon_loss: 0.295325, MMD_loss:3.642399, time: 0.238 sec\n",
+            "[115/200] loss: 0.295523 Recon_loss: 0.295163, MMD_loss:3.601054, time: 0.245 sec\n",
+            "[116/200] loss: 0.295666 Recon_loss: 0.295305, MMD_loss:3.606743, time: 0.239 sec\n",
+            "[117/200] loss: 0.295715 Recon_loss: 0.295347, MMD_loss:3.678063, time: 0.243 sec\n",
+            "[118/200] loss: 0.295649 Recon_loss: 0.295281, MMD_loss:3.674895, time: 0.263 sec\n",
+            "[119/200] loss: 0.295659 Recon_loss: 0.295300, MMD_loss:3.591405, time: 0.253 sec\n",
+            "[120/200] loss: 0.295461 Recon_loss: 0.295096, MMD_loss:3.652637, time: 0.247 sec\n",
+            "[121/200] loss: 0.295927 Recon_loss: 0.295572, MMD_loss:3.549883, time: 0.256 sec\n",
+            "[122/200] loss: 0.296130 Recon_loss: 0.295767, MMD_loss:3.629006, time: 0.255 sec\n",
+            "[123/200] loss: 0.296257 Recon_loss: 0.295894, MMD_loss:3.634080, time: 0.245 sec\n",
+            "[124/200] loss: 0.296348 Recon_loss: 0.295987, MMD_loss:3.613013, time: 0.250 sec\n",
+            "[125/200] loss: 0.295927 Recon_loss: 0.295578, MMD_loss:3.499508, time: 0.234 sec\n",
+            "[126/200] loss: 0.295687 Recon_loss: 0.295322, MMD_loss:3.647924, time: 0.249 sec\n",
+            "[127/200] loss: 0.295282 Recon_loss: 0.294926, MMD_loss:3.557264, time: 0.237 sec\n",
+            "[128/200] loss: 0.295794 Recon_loss: 0.295428, MMD_loss:3.664635, time: 0.241 sec\n",
+            "[129/200] loss: 0.295555 Recon_loss: 0.295195, MMD_loss:3.601915, time: 0.255 sec\n",
+            "[130/200] loss: 0.295537 Recon_loss: 0.295176, MMD_loss:3.612277, time: 0.254 sec\n",
+            "[131/200] loss: 0.295631 Recon_loss: 0.295279, MMD_loss:3.522908, time: 0.288 sec\n",
+            "[132/200] loss: 0.295648 Recon_loss: 0.295301, MMD_loss:3.469494, time: 0.267 sec\n",
+            "[133/200] loss: 0.295781 Recon_loss: 0.295428, MMD_loss:3.527443, time: 0.266 sec\n",
+            "[134/200] loss: 0.295831 Recon_loss: 0.295477, MMD_loss:3.538476, time: 0.256 sec\n",
+            "[135/200] loss: 0.295494 Recon_loss: 0.295145, MMD_loss:3.489057, time: 0.283 sec\n",
+            "[136/200] loss: 0.295453 Recon_loss: 0.295103, MMD_loss:3.500106, time: 0.251 sec\n",
+            "[137/200] loss: 0.295375 Recon_loss: 0.295030, MMD_loss:3.455212, time: 0.267 sec\n",
+            "[138/200] loss: 0.295315 Recon_loss: 0.294967, MMD_loss:3.476293, time: 0.275 sec\n",
+            "[139/200] loss: 0.295587 Recon_loss: 0.295232, MMD_loss:3.551312, time: 0.260 sec\n",
+            "[140/200] loss: 0.295609 Recon_loss: 0.295255, MMD_loss:3.542232, time: 0.255 sec\n",
+            "[141/200] loss: 0.295770 Recon_loss: 0.295419, MMD_loss:3.508347, time: 0.260 sec\n",
+            "[142/200] loss: 0.295343 Recon_loss: 0.295003, MMD_loss:3.404988, time: 0.243 sec\n",
+            "[143/200] loss: 0.295554 Recon_loss: 0.295213, MMD_loss:3.410123, time: 0.253 sec\n",
+            "[144/200] loss: 0.295452 Recon_loss: 0.295108, MMD_loss:3.432307, time: 0.245 sec\n",
+            "[145/200] loss: 0.295344 Recon_loss: 0.294997, MMD_loss:3.468390, time: 0.269 sec\n",
+            "[146/200] loss: 0.295619 Recon_loss: 0.295265, MMD_loss:3.541102, time: 0.256 sec\n",
+            "[147/200] loss: 0.295411 Recon_loss: 0.295063, MMD_loss:3.487115, time: 0.251 sec\n",
+            "[148/200] loss: 0.295695 Recon_loss: 0.295348, MMD_loss:3.464036, time: 0.241 sec\n",
+            "[149/200] loss: 0.295660 Recon_loss: 0.295312, MMD_loss:3.484148, time: 0.258 sec\n",
+            "[150/200] loss: 0.295666 Recon_loss: 0.295326, MMD_loss:3.399888, time: 0.244 sec\n",
+            "[151/200] loss: 0.295601 Recon_loss: 0.295256, MMD_loss:3.448928, time: 0.235 sec\n",
+            "[152/200] loss: 0.295677 Recon_loss: 0.295339, MMD_loss:3.381584, time: 0.249 sec\n",
+            "[153/200] loss: 0.295859 Recon_loss: 0.295513, MMD_loss:3.458268, time: 0.254 sec\n",
+            "[154/200] loss: 0.295812 Recon_loss: 0.295470, MMD_loss:3.419529, time: 0.255 sec\n",
+            "[155/200] loss: 0.295517 Recon_loss: 0.295166, MMD_loss:3.503078, time: 0.241 sec\n",
+            "[156/200] loss: 0.295478 Recon_loss: 0.295126, MMD_loss:3.517942, time: 0.254 sec\n",
+            "[157/200] loss: 0.295616 Recon_loss: 0.295273, MMD_loss:3.433700, time: 0.249 sec\n",
+            "[158/200] loss: 0.295544 Recon_loss: 0.295199, MMD_loss:3.445509, time: 0.239 sec\n",
+            "[159/200] loss: 0.295500 Recon_loss: 0.295156, MMD_loss:3.443838, time: 0.244 sec\n",
+            "[160/200] loss: 0.295503 Recon_loss: 0.295156, MMD_loss:3.471210, time: 0.252 sec\n",
+            "[161/200] loss: 0.295527 Recon_loss: 0.295175, MMD_loss:3.521035, time: 0.243 sec\n",
+            "[162/200] loss: 0.295851 Recon_loss: 0.295498, MMD_loss:3.529360, time: 0.235 sec\n",
+            "[163/200] loss: 0.295302 Recon_loss: 0.294957, MMD_loss:3.453178, time: 0.241 sec\n",
+            "[164/200] loss: 0.295388 Recon_loss: 0.295054, MMD_loss:3.341036, time: 0.255 sec\n",
+            "[165/200] loss: 0.295449 Recon_loss: 0.295109, MMD_loss:3.392378, time: 0.245 sec\n",
+            "[166/200] loss: 0.295816 Recon_loss: 0.295473, MMD_loss:3.429021, time: 0.246 sec\n",
+            "[167/200] loss: 0.295623 Recon_loss: 0.295275, MMD_loss:3.480336, time: 0.243 sec\n",
+            "[168/200] loss: 0.295421 Recon_loss: 0.295088, MMD_loss:3.329207, time: 0.236 sec\n",
+            "[169/200] loss: 0.295350 Recon_loss: 0.295013, MMD_loss:3.375183, time: 0.243 sec\n",
+            "[170/200] loss: 0.295432 Recon_loss: 0.295096, MMD_loss:3.361101, time: 0.252 sec\n",
+            "[171/200] loss: 0.295338 Recon_loss: 0.295008, MMD_loss:3.303618, time: 0.242 sec\n",
+            "[172/200] loss: 0.295325 Recon_loss: 0.294988, MMD_loss:3.371090, time: 0.257 sec\n",
+            "[173/200] loss: 0.295271 Recon_loss: 0.294933, MMD_loss:3.381522, time: 0.248 sec\n",
+            "[174/200] loss: 0.295407 Recon_loss: 0.295076, MMD_loss:3.302323, time: 0.266 sec\n",
+            "[175/200] loss: 0.295337 Recon_loss: 0.295008, MMD_loss:3.284310, time: 0.274 sec\n",
+            "[176/200] loss: 0.295164 Recon_loss: 0.294837, MMD_loss:3.272789, time: 0.259 sec\n",
+            "[177/200] loss: 0.295428 Recon_loss: 0.295099, MMD_loss:3.288365, time: 0.245 sec\n",
+            "[178/200] loss: 0.295433 Recon_loss: 0.295098, MMD_loss:3.353295, time: 0.231 sec\n",
+            "[179/200] loss: 0.295287 Recon_loss: 0.294956, MMD_loss:3.308256, time: 0.248 sec\n",
+            "[180/200] loss: 0.295213 Recon_loss: 0.294881, MMD_loss:3.322318, time: 0.237 sec\n",
+            "[181/200] loss: 0.295151 Recon_loss: 0.294819, MMD_loss:3.315457, time: 0.256 sec\n",
+            "[182/200] loss: 0.295405 Recon_loss: 0.295078, MMD_loss:3.266051, time: 0.237 sec\n",
+            "[183/200] loss: 0.295202 Recon_loss: 0.294871, MMD_loss:3.304138, time: 0.233 sec\n",
+            "[184/200] loss: 0.295198 Recon_loss: 0.294868, MMD_loss:3.307349, time: 0.238 sec\n",
+            "[185/200] loss: 0.295473 Recon_loss: 0.295132, MMD_loss:3.419493, time: 0.239 sec\n",
+            "[186/200] loss: 0.295252 Recon_loss: 0.294922, MMD_loss:3.294622, time: 0.253 sec\n",
+            "[187/200] loss: 0.295263 Recon_loss: 0.294924, MMD_loss:3.388800, time: 0.240 sec\n",
+            "[188/200] loss: 0.295256 Recon_loss: 0.294931, MMD_loss:3.248430, time: 0.230 sec\n",
+            "[189/200] loss: 0.295330 Recon_loss: 0.295001, MMD_loss:3.291317, time: 0.240 sec\n",
+            "[190/200] loss: 0.295174 Recon_loss: 0.294853, MMD_loss:3.210816, time: 0.263 sec\n",
+            "[191/200] loss: 0.295024 Recon_loss: 0.294706, MMD_loss:3.181574, time: 0.238 sec\n",
+            "[192/200] loss: 0.295310 Recon_loss: 0.294985, MMD_loss:3.251767, time: 0.240 sec\n",
+            "[193/200] loss: 0.295335 Recon_loss: 0.295004, MMD_loss:3.304505, time: 0.251 sec\n",
+            "[194/200] loss: 0.295181 Recon_loss: 0.294854, MMD_loss:3.269891, time: 0.246 sec\n",
+            "[195/200] loss: 0.295254 Recon_loss: 0.294930, MMD_loss:3.235049, time: 0.258 sec\n",
+            "[196/200] loss: 0.295156 Recon_loss: 0.294833, MMD_loss:3.236026, time: 0.252 sec\n",
+            "[197/200] loss: 0.295175 Recon_loss: 0.294856, MMD_loss:3.193847, time: 0.231 sec\n",
+            "[198/200] loss: 0.295324 Recon_loss: 0.294997, MMD_loss:3.265199, time: 0.247 sec\n",
+            "[199/200] loss: 0.295180 Recon_loss: 0.294861, MMD_loss:3.184954, time: 0.231 sec\n",
+            "[200/200] loss: 0.295205 Recon_loss: 0.294878, MMD_loss:3.269672, time: 0.247 sec\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "rSr013Mk7qeP"
+      },
+      "source": [
+        "**double check on the recontruted compositions**\n",
+        "\n",
+        "*   one way to find out whether WAE (or any other VAE) has learned the \n",
+        "repsentation is to compare the reconstructed and original compositions. \n",
+        "\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 234
+        },
+        "id": "vcz36-9atIfk",
+        "outputId": "33111178-fbc0-42a3-85c2-e454e56edd67"
+      },
+      "source": [
+        "#double check on the recontruted compositions\n",
+        "#t = time.localtime()\n",
+        "model_dir = os.path.join(root,'{}/{}_200.pth'.format(params['model_name'], params['model_name']))\n",
+        "model = WAE(raw_x.shape[1]).to(device)\n",
+        "model.load_state_dict(torch.load(model_dir))\n",
+        "model.eval()\n",
+        "with torch.no_grad():\n",
+        "    test = torch.FloatTensor(raw_x).to(device)\n",
+        "    recon_x, z = model(test)\n",
+        "    recon_x = model.decoder(z)\n",
+        "    recon_x = recon_x.cpu().detach().numpy()\n",
+        "\n",
+        "column_name = ['Fe','Ni','Co','Cr','V','Cu']#,'VEC','AR1','AR2','PE','Density','TC','MP','FI','SI','TI','M']\n",
+        "#recon_x = (recon_x * (max-min)) + min\n",
+        "pd.DataFrame(recon_x.round(3), columns=column_name).loc[690:695]"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>Fe</th>\n",
+              "      <th>Ni</th>\n",
+              "      <th>Co</th>\n",
+              "      <th>Cr</th>\n",
+              "      <th>V</th>\n",
+              "      <th>Cu</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>690</th>\n",
+              "      <td>0.646</td>\n",
+              "      <td>0.315</td>\n",
+              "      <td>0.038</td>\n",
+              "      <td>0.000</td>\n",
+              "      <td>0.001</td>\n",
+              "      <td>0.0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>691</th>\n",
+              "      <td>0.645</td>\n",
+              "      <td>0.316</td>\n",
+              "      <td>0.038</td>\n",
+              "      <td>0.000</td>\n",
+              "      <td>0.001</td>\n",
+              "      <td>0.0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>692</th>\n",
+              "      <td>0.641</td>\n",
+              "      <td>0.312</td>\n",
+              "      <td>0.046</td>\n",
+              "      <td>0.000</td>\n",
+              "      <td>0.001</td>\n",
+              "      <td>0.0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>693</th>\n",
+              "      <td>0.639</td>\n",
+              "      <td>0.314</td>\n",
+              "      <td>0.045</td>\n",
+              "      <td>0.000</td>\n",
+              "      <td>0.001</td>\n",
+              "      <td>0.0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>694</th>\n",
+              "      <td>0.639</td>\n",
+              "      <td>0.307</td>\n",
+              "      <td>0.053</td>\n",
+              "      <td>0.001</td>\n",
+              "      <td>0.000</td>\n",
+              "      <td>0.0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>695</th>\n",
+              "      <td>0.636</td>\n",
+              "      <td>0.309</td>\n",
+              "      <td>0.054</td>\n",
+              "      <td>0.000</td>\n",
+              "      <td>0.001</td>\n",
+              "      <td>0.0</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "        Fe     Ni     Co     Cr      V   Cu\n",
+              "690  0.646  0.315  0.038  0.000  0.001  0.0\n",
+              "691  0.645  0.316  0.038  0.000  0.001  0.0\n",
+              "692  0.641  0.312  0.046  0.000  0.001  0.0\n",
+              "693  0.639  0.314  0.045  0.000  0.001  0.0\n",
+              "694  0.639  0.307  0.053  0.001  0.000  0.0\n",
+              "695  0.636  0.309  0.054  0.000  0.001  0.0"
+            ]
+          },
+          "metadata": {},
+          "execution_count": 10
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 233
+        },
+        "id": "uIm7HRMYthG5",
+        "outputId": "e0bc43c5-5844-4a23-8435-6da0611dfa4a"
+      },
+      "source": [
+        "csv_data = pd.read_csv('data_base.csv', header=0).iloc[:,1:19]\n",
+        "csv_data.iloc[690:696,:6].round(3)"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>Fe</th>\n",
+              "      <th>Ni</th>\n",
+              "      <th>Co</th>\n",
+              "      <th>Cr</th>\n",
+              "      <th>V</th>\n",
+              "      <th>Cu</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>690</th>\n",
+              "      <td>0.630</td>\n",
+              "      <td>0.330</td>\n",
+              "      <td>0.04</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>691</th>\n",
+              "      <td>0.625</td>\n",
+              "      <td>0.335</td>\n",
+              "      <td>0.04</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>692</th>\n",
+              "      <td>0.635</td>\n",
+              "      <td>0.315</td>\n",
+              "      <td>0.05</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>693</th>\n",
+              "      <td>0.625</td>\n",
+              "      <td>0.325</td>\n",
+              "      <td>0.05</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>694</th>\n",
+              "      <td>0.635</td>\n",
+              "      <td>0.305</td>\n",
+              "      <td>0.06</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>695</th>\n",
+              "      <td>0.625</td>\n",
+              "      <td>0.315</td>\n",
+              "      <td>0.06</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.0</td>\n",
+              "      <td>0.0</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "        Fe     Ni    Co   Cr    V   Cu\n",
+              "690  0.630  0.330  0.04  0.0  0.0  0.0\n",
+              "691  0.625  0.335  0.04  0.0  0.0  0.0\n",
+              "692  0.635  0.315  0.05  0.0  0.0  0.0\n",
+              "693  0.625  0.325  0.05  0.0  0.0  0.0\n",
+              "694  0.635  0.305  0.06  0.0  0.0  0.0\n",
+              "695  0.625  0.315  0.06  0.0  0.0  0.0"
+            ]
+          },
+          "metadata": {},
+          "execution_count": 11
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "7uheLggK8WqX"
+      },
+      "source": [
+        "**Visualize the WAE latent space**\n",
+        "\n",
+        "Here we assign different colors to alloy with and without Copper,\n",
+        "as we expected them to differ significantly in the latent space.\n",
+        "\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 549
+        },
+        "id": "Ua13kwzjt0jv",
+        "outputId": "37f81f68-8054-4e60-ac8f-7751f7607ea4"
+      },
+      "source": [
+        "sns.set_style('ticks')\n",
+        "model = WAE(raw_x.shape[1]).to(device)\n",
+        "model.load_state_dict(torch.load(model_dir))\n",
+        "dataset = FeatureDataset(raw_x[:], raw_y[:])\n",
+        "latents = get_latents(model, dataset)\n",
+        "\n",
+        "low_cu = raw_x[:,5] < 0.05\n",
+        "low_cu_latent = latents[low_cu]\n",
+        "low_cu_color = raw_y[:][low_cu]\n",
+        "\n",
+        "high_cu = raw_x[:,5] >= 0.05\n",
+        "high_cu_latent = latents[high_cu]\n",
+        "high_cu_color = raw_y[:][high_cu]\n",
+        "\n",
+        "\n",
+        "# figure settings\n",
+        "fig, axs = plt.subplots(figsize = (3, 3),dpi=200)\n",
+        "\n",
+        "#axs.set_aspect(1.)\n",
+        "#axs.set_ylim(-7,7)\n",
+        "#axs.set_xlim(-11,5)\n",
+        "\n",
+        "axs.set_yticks(np.arange(-6, 8, step=2))\n",
+        "axs.set_xticks(np.arange(-10, 5, step=2))\n",
+        "\n",
+        "axs.set_yticklabels(np.arange(-6, 8, step=2), fontsize=7)\n",
+        "axs.set_xticklabels(np.arange(-10, 5, step=2), fontsize=7)\n",
+        "\n",
+        "\n",
+        "for axis in ['top','bottom','left','right']:\n",
+        "  axs.spines[axis].set_linewidth(1.)\n",
+        "\n",
+        "\n",
+        "axs.tick_params(axis='both', which='major', top=False, labeltop=False, direction='out', width=1., length=4)\n",
+        "axs.tick_params(axis='both', which='major', right=False, labelright=False, direction='out', width=1., length=4)\n",
+        "\n",
+        "#scatter1 = axs.scatter(low_cu_latent[:,0], low_cu_latent[:,1], c=low_cu_color, alpha=.75, s=10, linewidths=0, cmap='viridis')\n",
+        "#scatter2 = axs.scatter(high_cu_latent[:,0], high_cu_latent[:,1], c=high_cu_color, alpha=.75, s=9, linewidths=0, cmap='Reds')\n",
+        "\n",
+        "scatter1 = axs.scatter(low_cu_latent[:,0], low_cu_latent[:,1], c='steelblue', alpha=.55, s=8, linewidths=0, label='Alloys w/o Cu')\n",
+        "scatter2 = axs.scatter(high_cu_latent[:,0], high_cu_latent[:,1], c='firebrick', alpha=.65, s=14, linewidths=0, marker='^', label='Alloys w/ Cu')\n",
+        "#scatter3 = axs.scatter(latents_exp_4[:,0], latents_exp_4[:,1], alpha=1., s=10, linewidths=.75, edgecolors='darkslategray', facecolors='w')#, label='New FeCoNiCr HEAs')\n",
+        "#scatter4 = axs.scatter(latents_exp_5[:,0], latents_exp_5[:,1], alpha=1., s=16, linewidths=.75, edgecolors='darkred', facecolors='w',marker='^')#, label='New FeCoNiCrCu HEAs')\n",
+        "\n",
+        "handles,labels = axs.get_legend_handles_labels()\n",
+        "handles = handles[::1]\n",
+        "labels = labels[::1]\n",
+        "\n",
+        "legend_properties = {'size':7.5}\n",
+        "axs.legend(handles, labels, loc='upper right', bbox_to_anchor=(1.015,1.017), handletextpad=-0.3, frameon=False, prop=legend_properties)\n",
+        "#axs.legend(handles, labels, loc='upper left', bbox_to_anchor=(-0.045,1.017), handletextpad=-0.3, frameon=False, prop=legend_properties)\n",
+        "\n",
+        "#rect = patches.Rectangle((-19.4,15.0), 18, 4.5, linewidth=0,edgecolor=None,facecolor='k', alpha=0.03,linestyle=None,zorder=-10) #(0.2,15.4), 14, 4.1\n",
+        "#axs.add_patch(rect)\n",
+        "\n",
+        "fig.savefig('Figure3_a.tif', bbox_inches = 'tight', pad_inches=0.01)"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 600x600 with 1 Axes>"
+            ]
+          },
+          "metadata": {}
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "6kU0c7vv9bkg"
+      },
+      "source": [
+        "#**From WAE latent space to composition**\n",
+        "\n",
+        "In order to sample composition from the latent space. We turn the latent space into probability distribution via Gausian Mixture Model (GMM) \n",
+        "\n",
+        "\n",
+        "\n",
+        "\n",
+        "\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "oS7EEdLX03Mo"
+      },
+      "source": [
+        "#plotting functions\n",
+        "def draw_ellipse(position, covariance, ax=None, **kwargs):\n",
+        "    \"\"\"Draw an ellipse with a given position and covariance\"\"\"\n",
+        "    ax = ax or plt.gca()\n",
+        "    \n",
+        "    # Convert covariance to principal axes\n",
+        "    if covariance.shape == (2, 2):\n",
+        "        U, s, Vt = np.linalg.svd(covariance)\n",
+        "        angle = np.degrees(np.arctan2(U[1, 0], U[0, 0]))\n",
+        "        width, height = 2 * np.sqrt(s)\n",
+        "    else:\n",
+        "        angle = 0\n",
+        "        width, height = 2 * np.sqrt(covariance)\n",
+        "    \n",
+        "    # Draw the Ellipse\n",
+        "    for nsig in range(1, 4):\n",
+        "        ax.add_patch(Ellipse(position, nsig * width, nsig * height,\n",
+        "                             angle, **kwargs))\n",
+        "        \n",
+        "def plot_gmm(gm, X, label=True, ax=None):\n",
+        "    X= latents\n",
+        "    fig, axs = plt.subplots(1,1,figsize=(2,2),dpi=200)\n",
+        "    ax = axs or plt.gca()\n",
+        "    labels = gm.fit(X).predict(X)\n",
+        "    if label:\n",
+        "        low_cu = raw_x[:,5] < 0.05\n",
+        "        low_cu_latent = latents[low_cu]\n",
+        "        low_cu_color = raw_y[:][low_cu]\n",
+        "\n",
+        "        high_cu = raw_x[:,5] >= 0.05\n",
+        "        high_cu_latent = latents[high_cu]\n",
+        "        high_cu_color = raw_y[:][high_cu]\n",
+        "\n",
+        "        scatter1 = axs.scatter(low_cu_latent[:,0], low_cu_latent[:,1], c=low_cu_color, alpha=.65, s=8, linewidths=0, cmap='viridis')\n",
+        "        scatter2 = axs.scatter(high_cu_latent[:,0], high_cu_latent[:,1], c=high_cu_color, alpha=.65, s=14, linewidths=0, cmap='Reds', marker='^')\n",
+        "        #scatter3 = axs.scatter(latents[698:,0], latents[698:,1], alpha=1., s=10, linewidths=.75, edgecolors='k', facecolors='none')\n",
+        "    else:\n",
+        "        ax.scatter(X[:, 0], X[:, 1], s=5, zorder=2)\n",
+        "    ax.axis('equal')\n",
+        "    \n",
+        "    w_factor = 0.2 / gm.weights_.max()\n",
+        "    for pos, covar, w in zip(gm.means_, gm.covariances_, gm.weights_):\n",
+        "        draw_ellipse(pos, covar, alpha= 0.75*w * w_factor, facecolor='slategrey', zorder=-10)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "Mhbv_7whyr__"
+      },
+      "source": [
+        "Here the GMM is applied, you might wonder why 4 is chosen, the answer can be found below"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 431
+        },
+        "id": "lud3Guz13qvc",
+        "outputId": "24f2847d-19dd-4161-9c02-36bf28b9857e"
+      },
+      "source": [
+        "gm = GaussianMixture(n_components=4, random_state=0, init_params='kmeans').fit(latents) #plot a n_components v.s. Average negative log likelihood\n",
+        "print('Average negative log likelihood:', -1*gm.score(latents))\n",
+        "plot_gmm(gm, latents)"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Average negative log likelihood: 1.997225223049401\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 400x400 with 1 Axes>"
+            ]
+          },
+          "metadata": {}
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "iaK6ggga-iRF"
+      },
+      "source": [
+        "Using elbow method to find out the best # of components, the best number of cluster is either 4 or 5."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "P8_AGKU333z8",
+        "outputId": "57be67a8-a61f-41ce-a8f0-3e30e3ed8182"
+      },
+      "source": [
+        "scores=[] #using elbow method to find out the best # of components\n",
+        "for i in range(1,12):\n",
+        "  gm = GaussianMixture(n_components=i, random_state=0, init_params='kmeans').fit(latents)\n",
+        "  print('Average negative log likelihood:', -1*gm.score(latents))\n",
+        "  scores.append(-1*gm.score(latents))"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Average negative log likelihood: 2.395885633850088\n",
+            "Average negative log likelihood: 2.2357299041228\n",
+            "Average negative log likelihood: 2.151813013300802\n",
+            "Average negative log likelihood: 1.997225223049401\n",
+            "Average negative log likelihood: 1.8285200765222327\n",
+            "Average negative log likelihood: 1.8036536255451854\n",
+            "Average negative log likelihood: 1.7927436792221625\n",
+            "Average negative log likelihood: 1.7941665269689822\n",
+            "Average negative log likelihood: 1.7775483462620811\n",
+            "Average negative log likelihood: 1.7336035843338682\n",
+            "Average negative log likelihood: 1.66399637356645\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 268
+        },
+        "id": "ywT4MXG45OkC",
+        "outputId": "367e662d-fc3f-4a89-8176-920ee797d9d6"
+      },
+      "source": [
+        "import matplotlib.pyplot as plt\n",
+        "plt.figure()\n",
+        "plt.scatter(range(1,12), scores,color='red')\n",
+        "plt.plot(range(1,12),scores)\n",
+        "plt.show()"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {}
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "AINTm8aV-yqP"
+      },
+      "source": [
+        "#INVAR Classifier\n",
+        "\n",
+        "A simple neural network classifier that predicts INVAR based on composition. "
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "80SQeJDw5er_"
+      },
+      "source": [
+        "class Classifier(nn.Module): #a very simple classifer with large dropout. intuition here: as simple as possible, given that we only have 2d input\n",
+        "    def __init__(self):\n",
+        "        super(Classifier, self).__init__()\n",
+        "        self.fc = nn.Sequential(\n",
+        "            nn.Linear(2,8),\n",
+        "            nn.Dropout(0.5),\n",
+        "            nn.Linear(8,1),\n",
+        "            nn.Sigmoid()\n",
+        "        )\n",
+        "    \n",
+        "    def forward(self, x):\n",
+        "        return self.fc(x)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "Gf_Nrnjx_1s_"
+      },
+      "source": [
+        "**Classifier training**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "6ervUJMQ5W7N",
+        "outputId": "633582ed-d556-479d-c463-2622c9b6eb9e"
+      },
+      "source": [
+        "same_seeds(1)\n",
+        "\n",
+        "params['cls_bs'] = 16\n",
+        "params['cls_lr'] = 1e-4\n",
+        "params['cls_epoch'] = 200\n",
+        "params['num_fold'] = 5\n",
+        "\n",
+        "\n",
+        "params['label_y'] = np.where(raw_y<5, 1, 0)\n",
+        "params['latents'] = latents\n",
+        "\n",
+        "cls = Classifier().to(device)\n",
+        "opt = Adam(cls.parameters(), lr=params['cls_lr'], weight_decay=0.)\n",
+        "\n",
+        "\n",
+        "def training_Cls(model, optimizer, params):\n",
+        "    label_y = params['label_y']\n",
+        "    latents = params['latents']\n",
+        "    cls_epoch = params['cls_epoch']\n",
+        "\n",
+        "    kf = KFold(n_splits=params['num_fold'])\n",
+        "    train_acc = []\n",
+        "    test_acc = []\n",
+        "\n",
+        "    k=1\n",
+        "    for train, test in kf.split(latents):\n",
+        "        x_train, x_test, y_train, y_test = latents[train], latents[test], label_y[train], label_y[test]\n",
+        "        cls_dataset = AttributeDataset(x_train, y_train)\n",
+        "        cls_dataloader = DataLoader(cls_dataset, batch_size=params['cls_bs'], shuffle=True)\n",
+        "        cls_testDataset = AttributeDataset(x_test, y_test)\n",
+        "        cls_testDataloader = DataLoader(cls_testDataset, batch_size=cls_testDataset.__len__(), shuffle=False)\n",
+        "\n",
+        "\n",
+        "        for epoch in range(cls_epoch):\n",
+        "            t = time.time()\n",
+        "            total_loss = []\n",
+        "            total_acc = []\n",
+        "            cls.train()\n",
+        "            \n",
+        "            for i, data in enumerate(cls_dataloader):\n",
+        "                x = data[0].to(device)\n",
+        "                y = data[1].to(device)\n",
+        "                y_pred = cls(x)\n",
+        "                loss = F.binary_cross_entropy(y_pred, y)\n",
+        "                total_acc.append(torch.sum(torch.where(y_pred>=0.5,1,0) == y).detach().cpu().numpy())\n",
+        "                total_loss.append(loss.item())\n",
+        "\n",
+        "                opt.zero_grad()\n",
+        "                loss.backward()\n",
+        "                opt.step()\n",
+        "            \n",
+        "            #eval\n",
+        "            cls.eval()\n",
+        "            for test in cls_testDataloader:\n",
+        "                x = test[0].to(device)\n",
+        "                y = test[1].to(device)\n",
+        "                y_pred = cls(x)\n",
+        "                accuracy = torch.sum(torch.where(y_pred>=0.5,1,0) == y) / y_pred.size(0)\n",
+        "                test_loss = F.binary_cross_entropy(y_pred, y)\n",
+        "\n",
+        "            #print(f'[{epoch+1:03}/{cls_epoch}] loss:{sum(total_loss)/len(total_loss):.3f} test_loss:{test_loss.item():.3f} acc:{sum(total_acc)/cls_dataset.__len__():.3f} test_acc:{accuracy:.3f} time:{time.time()-t:.3f}')\n",
+        "        \n",
+        "        print('[{}/{}] train_acc: {:.04f} || test_acc: {:.04f}'.format(k, params['num_fold'], sum(total_acc)/cls_dataset.__len__(), accuracy.item()))\n",
+        "        train_acc.append(sum(total_acc)/cls_dataset.__len__())\n",
+        "        test_acc.append(accuracy.item())\n",
+        "        k+=1\n",
+        "    print('train_acc: {:.04f} || test_acc: {:.04f}'.format(sum(train_acc)/len(train_acc), sum(test_acc)/len(test_acc)))\n",
+        "    return train_acc, test_acc\n",
+        "\n",
+        "train_acc, test_acc = training_Cls(cls, opt, params)"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "[1/5] train_acc: 0.8327 || test_acc: 0.8929\n",
+            "[2/5] train_acc: 0.8743 || test_acc: 0.7122\n",
+            "[3/5] train_acc: 0.8187 || test_acc: 0.9065\n",
+            "[4/5] train_acc: 0.8528 || test_acc: 0.8058\n",
+            "[5/5] train_acc: 0.8294 || test_acc: 0.9137\n",
+            "train_acc: 0.8416 || test_acc: 0.8462\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "hmkRO4PTAbmE"
+      },
+      "source": [
+        "#Markov Chain Monte Carlo for composition sampling"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "tbBx1tu_6HR2"
+      },
+      "source": [
+        "def MCMC(gm, classifier, n_samples, sigma=0.1): #MCMC\n",
+        "    sample_z = []\n",
+        "\n",
+        "    z = gm.sample(1)[0]\n",
+        "    for i in range(n_samples):\n",
+        "        uniform_rand = np.random.uniform(size=1)\n",
+        "        z_next = np.random.multivariate_normal(z.squeeze(),sigma*np.eye(2)).reshape(1,-1)\n",
+        "\n",
+        "        z_combined = np.concatenate((z, z_next),axis=0)\n",
+        "        scores = cls(torch.Tensor(z_combined).to(device)).detach().cpu().numpy().squeeze() \n",
+        "        z_score, z_next_score = np.log(scores[0]), np.log(scores[1]) #z score needes to be converted to log, coz gm score is log.\n",
+        "        z_prob, z_next_prob = (gm.score(z)+z_score), (gm.score(z_next)+z_next_score) # two log addition, output: log probability\n",
+        "        accepence = min(0, (z_next_prob - z_prob))\n",
+        "\n",
+        "        if i == 0:\n",
+        "            sample_z.append(z.squeeze())\n",
+        "\n",
+        "        if np.log(uniform_rand) < accepence:\n",
+        "            sample_z.append(z_next.squeeze())\n",
+        "            z = z_next\n",
+        "        else:\n",
+        "            pass\n",
+        "\n",
+        "    return np.stack(sample_z)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "TGX6Y-taAp2v"
+      },
+      "source": [
+        "Sample 10000 times with sigma=0.5"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "Z4uB20Bf6VZ_",
+        "outputId": "51e27575-9aa8-43bd-fac8-21ebd79c984b"
+      },
+      "source": [
+        "sample_z = MCMC(gm=gm, classifier=cls, n_samples=10000, sigma=0.5)\n",
+        "WAE_comps = model._decode(torch.Tensor(sample_z).to(device)).detach().cpu().numpy()  # new_comps save as csv and goes to TERM\n",
+        "print('Sample size:', sample_z.shape)"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Sample size: (2966, 2)\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "YJJIWk8L6go6",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 203
+        },
+        "outputId": "0fb56e94-9c50-411e-faa2-899fbfba91be"
+      },
+      "source": [
+        "WAE_comps=pd.DataFrame(WAE_comps)\n",
+        "WAE_comps.columns=column_name\n",
+        "WAE_comps.to_csv('comps_WAE.csv',index=False)\n",
+        "WAE_comps.head()"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>Fe</th>\n",
+              "      <th>Ni</th>\n",
+              "      <th>Co</th>\n",
+              "      <th>Cr</th>\n",
+              "      <th>V</th>\n",
+              "      <th>Cu</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>0.014077</td>\n",
+              "      <td>0.231521</td>\n",
+              "      <td>0.754396</td>\n",
+              "      <td>1.664660e-06</td>\n",
+              "      <td>0.000004</td>\n",
+              "      <td>1.433623e-07</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>0.141357</td>\n",
+              "      <td>0.333938</td>\n",
+              "      <td>0.524631</td>\n",
+              "      <td>7.136076e-07</td>\n",
+              "      <td>0.000072</td>\n",
+              "      <td>3.349655e-07</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>0.271059</td>\n",
+              "      <td>0.108174</td>\n",
+              "      <td>0.540923</td>\n",
+              "      <td>8.701057e-07</td>\n",
+              "      <td>0.079841</td>\n",
+              "      <td>1.835431e-06</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>0.274571</td>\n",
+              "      <td>0.086473</td>\n",
+              "      <td>0.587006</td>\n",
+              "      <td>1.669761e-06</td>\n",
+              "      <td>0.051947</td>\n",
+              "      <td>1.217349e-06</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>0.461331</td>\n",
+              "      <td>0.259892</td>\n",
+              "      <td>0.183169</td>\n",
+              "      <td>2.289097e-06</td>\n",
+              "      <td>0.095604</td>\n",
+              "      <td>2.271647e-06</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "         Fe        Ni        Co            Cr         V            Cu\n",
+              "0  0.014077  0.231521  0.754396  1.664660e-06  0.000004  1.433623e-07\n",
+              "1  0.141357  0.333938  0.524631  7.136076e-07  0.000072  3.349655e-07\n",
+              "2  0.271059  0.108174  0.540923  8.701057e-07  0.079841  1.835431e-06\n",
+              "3  0.274571  0.086473  0.587006  1.669761e-06  0.051947  1.217349e-06\n",
+              "4  0.461331  0.259892  0.183169  2.289097e-06  0.095604  2.271647e-06"
+            ]
+          },
+          "metadata": {},
+          "execution_count": 26
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "lwnDaCQmA72h"
+      },
+      "source": [
+        "Plotting the sampled composition along with known datas"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 565
+        },
+        "id": "VlhEqqLS8N9x",
+        "outputId": "d2d42a7b-5247-4a28-b61c-aef817d01dc1"
+      },
+      "source": [
+        "sns.set_style('ticks')\n",
+        "\n",
+        "t = time.localtime()\n",
+        "model_dir = os.path.join(root, '{}/{}_100.pth'.format(params['model_name'], params['model_name']))\n",
+        "model = WAE(raw_x.shape[1]).to(device)\n",
+        "model.load_state_dict(torch.load(model_dir))\n",
+        "model.eval()\n",
+        "\n",
+        "dataset = FeatureDataset(raw_x[:], raw_y[:])\n",
+        "latents = get_latents(model,dataset)\n",
+        "\n",
+        "low_cu = raw_x[:,5] < 0.05\n",
+        "low_cu_latent = latents[low_cu]\n",
+        "low_cu_color = raw_y[:][low_cu]\n",
+        "\n",
+        "high_cu = raw_x[:,5] >= 0.05\n",
+        "high_cu_latent = latents[high_cu]\n",
+        "high_cu_color = raw_y[:][high_cu]\n",
+        "\n",
+        "fig, axs = plt.subplots(figsize = (3, 3),dpi=200)\n",
+        "\n",
+        "axs.set_xlim(-6,3)\n",
+        "axs.set_ylim(-4,5)\n",
+        "\n",
+        "scatter1 = axs.scatter(low_cu_latent[:,0], low_cu_latent[:,1], c='steelblue', alpha=.55, s=2, linewidths=0, cmap='viridis')\n",
+        "scatter2 = axs.scatter(high_cu_latent[:,0], high_cu_latent[:,1], c=high_cu_color, alpha=.65, s=3.5, linewidths=0, cmap='Reds', marker='^')\n",
+        "\n",
+        "scatter4 = axs.scatter(sample_z[:,0], sample_z[:,1], c='k', alpha=.15, s=0.8, linewidths=0, zorder=-1)\n",
+        "\n",
+        "plt.show()"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 600x600 with 1 Axes>"
+            ]
+          },
+          "metadata": {}
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "jKkdRjD1DZQN"
+      },
+      "source": [
+        "#Two-stage ensemble regression model (TERM)"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "Cs6dt9E0ELs9"
+      },
+      "source": [
+        "**Ultility functions**\n",
+        "*   MAPELoss - Mean Absolute Percentile Error (MAPE)\n",
+        "*   normalizing_data - Normalizing the data\n",
+        "\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "k71dxvuL84R6"
+      },
+      "source": [
+        "import torch.nn as nn\n",
+        "import torch\n",
+        "import numpy as np\n",
+        "import pandas as pd\n",
+        "from torch.utils.data import Dataset, DataLoader\n",
+        "import random\n",
+        "\n",
+        "class MAPELoss(nn.Module):\n",
+        "     def __init__(self):\n",
+        "        super(MAPELoss, self).__init__() \n",
+        "        \n",
+        "     def forward (self, output, target):\n",
+        "         loss = torch.mean(torch.abs((target - output) / target))\n",
+        "         # loss = (-1)*loss\n",
+        "         return loss\n",
+        "     \n",
+        "def minmaxscaler(data):\n",
+        "    min = np.amin(data)\n",
+        "    max = np.amax(data)    \n",
+        "    return (data - min)/(max-min)\n",
+        "\n",
+        "def weights_init(m):\n",
+        "    classname = m.__class__.__name__\n",
+        "    if classname.find('BatchNorm') != -1:\n",
+        "        m.weight.data.normal_(1.0, 0.02)\n",
+        "        m.bias.data.fill_(0)\n",
+        "        \n",
+        "class FeatureDataset(Dataset):\n",
+        "    '''\n",
+        "    Args: x is a 2D numpy array [x_size, x_features]\n",
+        "    '''\n",
+        "    def __init__(self, x):\n",
+        "        self.x = x\n",
+        "    \n",
+        "    def __len__(self):\n",
+        "        return self.x.shape[0]\n",
+        "    \n",
+        "    def __getitem__(self, idx):\n",
+        "        return torch.FloatTensor(self.x[idx])\n",
+        "\n",
+        "    def getBatch(self, idxs = []):\n",
+        "        if idxs == None:\n",
+        "            return idxs\n",
+        "        else:\n",
+        "            x_features = []\n",
+        "            for i in idxs:\n",
+        "                x_features.append(self.__getitem__(i))\n",
+        "            return torch.FloatTensor(x_features)\n",
+        "        \n",
+        "def normalizing_data(data, seed=42):  \n",
+        "  df_all = data.drop(columns=['alloy'])\n",
+        "  #create a min max processing object\n",
+        "  composition = df_all [['Fe','Ni','Co','Cr','V','Cu']]\n",
+        "  min_max_scaler = preprocessing.MinMaxScaler()\n",
+        "  normalized_atomic_properties = min_max_scaler.fit_transform(df_all[['VEC','AR1','AR2','PE','Density',\n",
+        "                                              'TC','MP','FI','SI','TI','M']])\n",
+        "  x = pd.concat([composition,pd.DataFrame(normalized_atomic_properties)],axis=1)\n",
+        "  x=x.iloc[:697]\n",
+        "  y = df_all[['TEC']][:697]\n",
+        "  # bins     = [18,35,48,109,202,234,525,687,695]\n",
+        "  bins     = [18,35,48,109,202,234,525,687]\n",
+        "  y_binned = np.digitize(y.index, bins, right=True) #stratified 7-fold: each folder contains a specific type of alloys (7 types in total, each takes 85% and 15% as training and testing)\n",
+        "\n",
+        "  x = torch.FloatTensor(x.values) #numpy to tensor\n",
+        "  y = torch.FloatTensor(y.values) #numpy to tensor\n",
+        "\n",
+        "  if torch.cuda.is_available():\n",
+        "      x = x.cuda()\n",
+        "      y = y.cuda() \n",
+        "  \n",
+        "  train_features, test_features, train_labels, test_labels = train_test_split(x, y, test_size=0.15, random_state=seed, stratify=y_binned)\n",
+        "  return train_features, test_features, train_labels, test_labels"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "dcW5DVUsEo0o"
+      },
+      "source": [
+        "**Data loading**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "h9GqV60vATsC"
+      },
+      "source": [
+        "import datetime\n",
+        "import torch.utils.data as Data\n",
+        "import pandas as pd\n",
+        "import torch\n",
+        "import torch.nn.functional as F    \n",
+        "import matplotlib.pyplot as plt\n",
+        "import numpy as np\n",
+        "from sklearn.model_selection import train_test_split\n",
+        "import torch.nn as nn\n",
+        "import torch.optim as optim\n",
+        "from bayes_opt import BayesianOptimization\n",
+        "import time\n",
+        "import os\n",
+        "from sklearn import preprocessing\n",
+        "\n",
+        "\n",
+        "t = time.localtime()\n",
+        "    \n",
+        "table = pd.DataFrame(columns=['target','batch_size','lr','module__n_hidden','module__w'])\n",
+        "\n",
+        "plt.close('all')\n",
+        "starttime = datetime.datetime.now()\n",
+        "data = pd.read_csv('data_base.csv')\n",
+        "\n",
+        "\n",
+        "train_features, test_features, train_labels, test_labels = normalizing_data(data, seed=42)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "IuRdKijDE01B",
+        "outputId": "f047e8ab-a7b5-4bc5-f20e-cc8792b763db"
+      },
+      "source": [
+        "  print(torch.cuda.is_available())"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "False\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "cqhFxU_yE3pc"
+      },
+      "source": [
+        "**Neural network architecture**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "g2TdhVbt9j5F"
+      },
+      "source": [
+        "\n",
+        "class Net(nn.Module):  \n",
+        "    def __init__(self, n_feature=17, n_hidden=218, n_output=1, w = 6):\n",
+        "        super(Net, self).__init__()    \n",
+        "        # self.BN=torch.nn.BatchNorm1d(n_hidden)\n",
+        "        self.hidden1 = torch.nn.Linear(n_feature, n_hidden) \n",
+        "        nn.init.kaiming_normal_(self.hidden1.weight)\n",
+        "        \n",
+        "        self.hiddens = nn.ModuleList ([nn.Linear(n_hidden, n_hidden) for i in range(w)])                            \n",
+        "        for m in self.hiddens:\n",
+        "            nn.init.kaiming_normal_(m.weight)   \n",
+        "        \n",
+        "        self.predict = torch.nn.Linear(n_hidden, n_output)  \n",
+        "        nn.init.kaiming_normal_(self.predict.weight)\n",
+        "\n",
+        "    def forward(self, x):  \n",
+        "        x = self.hidden1(x)\n",
+        "        # x = self.BN(x)\n",
+        "        # x = self.Dropout (x)\n",
+        "        x = F.relu(x)   \n",
+        "        \n",
+        "        for m in self.hiddens:\n",
+        "            x = m(x)\n",
+        "            # x = self.BN(x)\n",
+        "            x = F.relu(x) \n",
+        "                      \n",
+        "        x = self.predict(x)\n",
+        "        # x = self.BN_3(x)\n",
+        "        # x = self.Dropout (x)\n",
+        "          # 输出值\n",
+        "        return x\n",
+        "\n",
+        "def train(net, num_epochs, batch_size, train_features, test_features, train_labels, test_labels,\n",
+        "          train_loader,\n",
+        "          optimizer):\n",
+        "    print (\"\\n=== train begin ===\")\n",
+        "    print(net)\n",
+        "    train_ls, test_ls = [], []\n",
+        "    loss = MAPELoss() # MAPE means Mean Absolute percentile error \n",
+        "    for epoch in range(num_epochs):\n",
+        "        for x, y in train_loader:\n",
+        "            ls = loss(net(x).view(-1, 1), y.view(-1, 1))\n",
+        "            optimizer.zero_grad()\n",
+        "            ls.backward()\n",
+        "            optimizer.step()\n",
+        "        if epoch % 100 == 0:\n",
+        "            train_ls.append(loss(net(train_features).view(-1, 1), train_labels.view(-1, 1)).item())\n",
+        "            test_ls.append(loss(net(test_features).view(-1, 1), test_labels.view(-1, 1)).item())\n",
+        "            print (\"epoch %d: train loss %f, test loss %f\" % (epoch, train_ls[-1], test_ls[-1]))\n",
+        "        \n",
+        "    print (\"=== train end ===\")\n",
+        "    \n",
+        "def test(model, test_loader):\n",
+        "    model.eval()\n",
+        "    test_loss = 0\n",
+        "    n = 0\n",
+        "    loss = MAPELoss() \n",
+        "    with torch.no_grad():\n",
+        "        for data, target in test_loader:\n",
+        "            output = model(data)\n",
+        "            test_loss += loss(output.view(-1, 1), target.view(-1, 1)).item()  # sum up batch loss\n",
+        "            n += 1\n",
+        "\n",
+        "    test_loss /= n\n",
+        "    \n",
+        "    print('Test set: Average loss: {:.4f}'.format(\n",
+        "        test_loss))\n",
+        "    \n",
+        "    return test_loss   \n",
+        "\n",
+        "    \n",
+        "\n",
+        "\n",
+        "def train_model(batch_size,lr, module__n_hidden,module__w):\n",
+        "    module__n_hidden = int(module__n_hidden) # number of neurons per layer\n",
+        "    module__w = int(module__w) # number of hidden layers\n",
+        "    batch_size = int(batch_size)\n",
+        "    train_dataset = Data.TensorDataset(train_features, train_labels)\n",
+        "    test_dataset = Data.TensorDataset(test_features, test_labels)\n",
+        "    train_loader = Data.DataLoader(train_dataset, batch_size, shuffle=True)\n",
+        "    test_loader = Data.DataLoader(test_dataset, batch_size, shuffle=True) \n",
+        "    net = Net(n_feature=17, n_hidden=module__n_hidden, n_output=1, w = module__w)\n",
+        "    if torch.cuda.is_available():\n",
+        "      net = net.cuda()\n",
+        "    n_epochs = 20 \n",
+        "    optimizer = optim.Adam(net.parameters(), lr=lr, weight_decay=0.0001)\n",
+        "    train(net, n_epochs, batch_size,train_features, test_features, \n",
+        "          train_labels, test_labels,train_loader, optimizer)\n",
+        "    train_loss= test(net,train_loader)\n",
+        "    test_loss = test(net, test_loader)\n",
+        "\n",
+        "    \n",
+        "    r = -np.abs(train_loss-test_loss)\n",
+        "    \n",
+        "    return -test_loss\n",
+        "          \n",
+        "\n",
+        "    "
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "FyzNkXLkFQXT"
+      },
+      "source": [
+        "**Bayesian hyperparameter optimization**\n",
+        "\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "bHuElmYp-2Ws",
+        "outputId": "5cf1139c-aeab-45cc-8a64-c135ca09f3cb"
+      },
+      "source": [
+        "bounds = {'lr': (0.0005,0.001), 'batch_size': (32,64), 'module__n_hidden': (16,526),\n",
+        "          'module__w': (2,10)}\n",
+        "optimizer = BayesianOptimization(\n",
+        "    f=train_model,\n",
+        "    pbounds=bounds,\n",
+        "    random_state=1,\n",
+        ")\n",
+        "\n",
+        "optimizer.maximize(init_points=10, n_iter=1)\n",
+        "print(optimizer.max)\n",
+        "table = pd.DataFrame(columns=['target','batch_size','lr','module__n_hidden','module__w'])\n",
+        "for res in optimizer.res:\n",
+        "    table=table.append(pd.DataFrame({'target':[res['target']],'batch_size':[res['params']['batch_size']],\n",
+        "                                     'lr':[res['params']['lr']], 'module__n_hidden':[res['params']['module__n_hidden']],\n",
+        "                                     'module__w':[res['params']['module__w']]}),ignore_index=True)\n",
+        "table=table.append(pd.DataFrame({'target':[optimizer.max['target']],'batch_size':[optimizer.max['params']['batch_size']],\n",
+        "                                    'lr':[optimizer.max['params']['lr']], 'module__n_hidden':[optimizer.max['params']['module__n_hidden']],\n",
+        "                                    'module__w':[optimizer.max['params']['module__w']]}),ignore_index=True)\n",
+        "model_name = 'Invar_inference_NN'\n",
+        "file_name = '{}.xlsx'.format(model_name)\n",
+        "endtime = datetime.datetime.now()\n",
+        "Rtime = endtime - starttime\n",
+        "print(Rtime)\n",
+        "table.to_excel(file_name)"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "|   iter    |  target   | batch_... |    lr     | module... | module__w |\n",
+            "-------------------------------------------------------------------------\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=16, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=16, out_features=16, bias=True)\n",
+            "    (1): Linear(in_features=16, out_features=16, bias=True)\n",
+            "    (2): Linear(in_features=16, out_features=16, bias=True)\n",
+            "    (3): Linear(in_features=16, out_features=16, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=16, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 1.090410, test loss 1.059720\n",
+            "=== train end ===\n",
+            "Test set: Average loss: 0.8703\n",
+            "Test set: Average loss: 0.9424\n",
+            "| \u001b[0m 1       \u001b[0m | \u001b[0m-0.9424  \u001b[0m | \u001b[0m 45.34   \u001b[0m | \u001b[0m 0.000860\u001b[0m | \u001b[0m 16.06   \u001b[0m | \u001b[0m 4.419   \u001b[0m |\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=110, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (1): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (2): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (3): Linear(in_features=110, out_features=110, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=110, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 0.961760, test loss 0.938732\n",
+            "=== train end ===\n",
+            "Test set: Average loss: 0.2994\n",
+            "Test set: Average loss: 0.3272\n",
+            "| \u001b[95m 2       \u001b[0m | \u001b[95m-0.3272  \u001b[0m | \u001b[95m 36.7    \u001b[0m | \u001b[95m 0.000546\u001b[0m | \u001b[95m 111.0   \u001b[0m | \u001b[95m 4.764   \u001b[0m |\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=229, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (1): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (2): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (3): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (4): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (5): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (6): Linear(in_features=229, out_features=229, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=229, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 0.949415, test loss 0.948496\n",
+            "=== train end ===\n",
+            "Test set: Average loss: 0.3680\n",
+            "Test set: Average loss: 0.3848\n",
+            "| \u001b[0m 3       \u001b[0m | \u001b[0m-0.3848  \u001b[0m | \u001b[0m 44.7    \u001b[0m | \u001b[0m 0.000769\u001b[0m | \u001b[0m 229.8   \u001b[0m | \u001b[0m 7.482   \u001b[0m |\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=29, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=29, out_features=29, bias=True)\n",
+            "    (1): Linear(in_features=29, out_features=29, bias=True)\n",
+            "    (2): Linear(in_features=29, out_features=29, bias=True)\n",
+            "    (3): Linear(in_features=29, out_features=29, bias=True)\n",
+            "    (4): Linear(in_features=29, out_features=29, bias=True)\n",
+            "    (5): Linear(in_features=29, out_features=29, bias=True)\n",
+            "    (6): Linear(in_features=29, out_features=29, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=29, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 0.969080, test loss 0.968986\n",
+            "=== train end ===\n",
+            "Test set: Average loss: 0.4092\n",
+            "Test set: Average loss: 0.4524\n",
+            "| \u001b[0m 4       \u001b[0m | \u001b[0m-0.4524  \u001b[0m | \u001b[0m 38.54   \u001b[0m | \u001b[0m 0.000939\u001b[0m | \u001b[0m 29.97   \u001b[0m | \u001b[0m 7.364   \u001b[0m |\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=87, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=87, out_features=87, bias=True)\n",
+            "    (1): Linear(in_features=87, out_features=87, bias=True)\n",
+            "    (2): Linear(in_features=87, out_features=87, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=87, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 0.975493, test loss 0.926396\n",
+            "=== train end ===\n",
+            "Test set: Average loss: 0.5092\n",
+            "Test set: Average loss: 0.4794\n",
+            "| \u001b[0m 5       \u001b[0m | \u001b[0m-0.4794  \u001b[0m | \u001b[0m 45.35   \u001b[0m | \u001b[0m 0.000779\u001b[0m | \u001b[0m 87.6    \u001b[0m | \u001b[0m 3.585   \u001b[0m |\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=175, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=175, out_features=175, bias=True)\n",
+            "    (1): Linear(in_features=175, out_features=175, bias=True)\n",
+            "    (2): Linear(in_features=175, out_features=175, bias=True)\n",
+            "    (3): Linear(in_features=175, out_features=175, bias=True)\n",
+            "    (4): Linear(in_features=175, out_features=175, bias=True)\n",
+            "    (5): Linear(in_features=175, out_features=175, bias=True)\n",
+            "    (6): Linear(in_features=175, out_features=175, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=175, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 0.960191, test loss 0.977315\n",
+            "=== train end ===\n",
+            "Test set: Average loss: 0.3248\n",
+            "Test set: Average loss: 0.3707\n",
+            "| \u001b[0m 6       \u001b[0m | \u001b[0m-0.3707  \u001b[0m | \u001b[0m 57.62   \u001b[0m | \u001b[0m 0.000984\u001b[0m | \u001b[0m 175.8   \u001b[0m | \u001b[0m 7.539   \u001b[0m |\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=59, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=59, out_features=59, bias=True)\n",
+            "    (1): Linear(in_features=59, out_features=59, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=59, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 0.972694, test loss 0.976538\n",
+            "=== train end ===\n",
+            "Test set: Average loss: 0.6441\n",
+            "Test set: Average loss: 0.6379\n",
+            "| \u001b[0m 7       \u001b[0m | \u001b[0m-0.6379  \u001b[0m | \u001b[0m 60.04   \u001b[0m | \u001b[0m 0.000947\u001b[0m | \u001b[0m 59.37   \u001b[0m | \u001b[0m 2.312   \u001b[0m |\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=66, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=66, out_features=66, bias=True)\n",
+            "    (1): Linear(in_features=66, out_features=66, bias=True)\n",
+            "    (2): Linear(in_features=66, out_features=66, bias=True)\n",
+            "    (3): Linear(in_features=66, out_features=66, bias=True)\n",
+            "    (4): Linear(in_features=66, out_features=66, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=66, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 0.955202, test loss 0.944204\n",
+            "=== train end ===\n",
+            "Test set: Average loss: 0.3059\n",
+            "Test set: Average loss: 0.3335\n",
+            "| \u001b[0m 8       \u001b[0m | \u001b[0m-0.3335  \u001b[0m | \u001b[0m 37.43   \u001b[0m | \u001b[0m 0.000939\u001b[0m | \u001b[0m 66.16   \u001b[0m | \u001b[0m 5.369   \u001b[0m |\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=368, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=368, out_features=368, bias=True)\n",
+            "    (1): Linear(in_features=368, out_features=368, bias=True)\n",
+            "    (2): Linear(in_features=368, out_features=368, bias=True)\n",
+            "    (3): Linear(in_features=368, out_features=368, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=368, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 0.902503, test loss 0.891311\n",
+            "=== train end ===\n",
+            "Test set: Average loss: 0.4176\n",
+            "Test set: Average loss: 0.4810\n",
+            "| \u001b[0m 9       \u001b[0m | \u001b[0m-0.481   \u001b[0m | \u001b[0m 62.65   \u001b[0m | \u001b[0m 0.000766\u001b[0m | \u001b[0m 368.9   \u001b[0m | \u001b[0m 4.524   \u001b[0m |\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=25, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=25, out_features=25, bias=True)\n",
+            "    (1): Linear(in_features=25, out_features=25, bias=True)\n",
+            "    (2): Linear(in_features=25, out_features=25, bias=True)\n",
+            "    (3): Linear(in_features=25, out_features=25, bias=True)\n",
+            "    (4): Linear(in_features=25, out_features=25, bias=True)\n",
+            "    (5): Linear(in_features=25, out_features=25, bias=True)\n",
+            "    (6): Linear(in_features=25, out_features=25, bias=True)\n",
+            "    (7): Linear(in_features=25, out_features=25, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=25, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 1.018245, test loss 1.006838\n",
+            "=== train end ===\n",
+            "Test set: Average loss: 0.7963\n",
+            "Test set: Average loss: 0.8304\n",
+            "| \u001b[0m 10      \u001b[0m | \u001b[0m-0.8304  \u001b[0m | \u001b[0m 53.97   \u001b[0m | \u001b[0m 0.000917\u001b[0m | \u001b[0m 25.33   \u001b[0m | \u001b[0m 8.001   \u001b[0m |\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=109, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=109, out_features=109, bias=True)\n",
+            "    (1): Linear(in_features=109, out_features=109, bias=True)\n",
+            "    (2): Linear(in_features=109, out_features=109, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=109, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 0.970867, test loss 0.911994\n",
+            "=== train end ===\n",
+            "Test set: Average loss: 0.3724\n",
+            "Test set: Average loss: 0.4202\n",
+            "| \u001b[0m 11      \u001b[0m | \u001b[0m-0.4202  \u001b[0m | \u001b[0m 37.32   \u001b[0m | \u001b[0m 0.000629\u001b[0m | \u001b[0m 109.1   \u001b[0m | \u001b[0m 3.425   \u001b[0m |\n",
+            "=========================================================================\n",
+            "{'target': -0.3271511296431224, 'params': {'batch_size': 36.69618850614762, 'lr': 0.0005461692973843989, 'module__n_hidden': 110.99270780261216, 'module__w': 4.764485816344382}}\n",
+            "0:04:57.421579\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 203
+        },
+        "id": "FE1V_s8QD-wv",
+        "outputId": "2f52ec0e-4790-4c28-c524-d83bd16f634f"
+      },
+      "source": [
+        "table.head()"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>target</th>\n",
+              "      <th>batch_size</th>\n",
+              "      <th>lr</th>\n",
+              "      <th>module__n_hidden</th>\n",
+              "      <th>module__w</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>-0.942386</td>\n",
+              "      <td>45.344704</td>\n",
+              "      <td>0.000860</td>\n",
+              "      <td>16.058331</td>\n",
+              "      <td>4.418661</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>-0.327151</td>\n",
+              "      <td>36.696189</td>\n",
+              "      <td>0.000546</td>\n",
+              "      <td>110.992708</td>\n",
+              "      <td>4.764486</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>-0.384802</td>\n",
+              "      <td>44.696559</td>\n",
+              "      <td>0.000769</td>\n",
+              "      <td>229.789202</td>\n",
+              "      <td>7.481756</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>-0.452406</td>\n",
+              "      <td>38.542472</td>\n",
+              "      <td>0.000939</td>\n",
+              "      <td>29.967673</td>\n",
+              "      <td>7.363740</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>-0.479412</td>\n",
+              "      <td>45.353754</td>\n",
+              "      <td>0.000779</td>\n",
+              "      <td>87.597339</td>\n",
+              "      <td>3.584812</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "     target  batch_size        lr  module__n_hidden  module__w\n",
+              "0 -0.942386   45.344704  0.000860         16.058331   4.418661\n",
+              "1 -0.327151   36.696189  0.000546        110.992708   4.764486\n",
+              "2 -0.384802   44.696559  0.000769        229.789202   7.481756\n",
+              "3 -0.452406   38.542472  0.000939         29.967673   7.363740\n",
+              "4 -0.479412   45.353754  0.000779         87.597339   3.584812"
+            ]
+          },
+          "metadata": {},
+          "execution_count": 45
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "2mqpUtXeGX47"
+      },
+      "source": [
+        "**Load top 2 NN model with 2 different seeds**\n",
+        "\n",
+        "Due to the data size problem, we even need to train the model with different seed in order to obtain a robust ensemble."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 1000
+        },
+        "id": "aZ7g7ajfD9fu",
+        "outputId": "4f4bc239-5f2d-4dc8-f3cb-a8c47f630058"
+      },
+      "source": [
+        "import datetime\n",
+        "import torch.utils.data as Data\n",
+        "import pandas as pd\n",
+        "import torch\n",
+        "import torch.nn.functional as F   \n",
+        "import matplotlib.pyplot as plt\n",
+        "import numpy as np\n",
+        "from sklearn.model_selection import train_test_split\n",
+        "import torch.nn as nn\n",
+        "import torch.optim as optim\n",
+        "import time\n",
+        "import os\n",
+        "import pickle \n",
+        "import seaborn as sns\n",
+        "t = time.localtime() \n",
+        "plt.close('all')\n",
+        "target = pd.read_excel('Invar_inference_NN.xlsx')\n",
+        "starttime = datetime.datetime.now()\n",
+        "for i in range(1,3): # This is to choose best 10 points \n",
+        "    for j in range(40,42): # 10 different seeds\n",
+        "        train_features, test_features, train_labels, test_labels = normalizing_data(data, seed=j)\n",
+        "        lr = target.at[i,'lr'] # the same\n",
+        "        module__n_hidden = target.at[i,'module__n_hidden']\n",
+        "        module__w = target.at[i,'module__w']\n",
+        "        batch_size = target.at[i,'batch_size']\n",
+        "        \n",
+        "        module__n_hidden = int(module__n_hidden)\n",
+        "        module__w = int(module__w)\n",
+        "        batch_size = int(batch_size)\n",
+        "        print (module__w)\n",
+        "        \n",
+        "        batch_size = target.at[i,'batch_size'] # choose 'batch_size' paramter at ith row\n",
+        "        lr = target.at[i,'lr'] # the same\n",
+        "        module__n_hidden = target.at[i,'module__n_hidden']\n",
+        "        module__w = target.at[i,'module__w']\n",
+        "        \n",
+        "        module__n_hidden = int(module__n_hidden)\n",
+        "        module__w = int(module__w)\n",
+        "        batch_size = int(batch_size)\n",
+        "        print (module__w)\n",
+        "        train_dataset = Data.TensorDataset(train_features, train_labels)\n",
+        "        test_dataset = Data.TensorDataset(test_features, test_labels)\n",
+        "        train_loader = Data.DataLoader(train_dataset, batch_size, shuffle=True)\n",
+        "        test_loader = Data.DataLoader(test_dataset, batch_size, shuffle=True) \n",
+        "            \n",
+        "\n",
+        "        class Net(nn.Module):  \n",
+        "            def __init__(self, n_feature=17, n_hidden=module__n_hidden, n_output=1, w = module__w):\n",
+        "                super(Net, self).__init__()   \n",
+        "                # self.BN=torch.nn.BatchNorm1d(n_hidden)\n",
+        "                self.hidden1 = torch.nn.Linear(n_feature, n_hidden) \n",
+        "                nn.init.kaiming_normal_(self.hidden1.weight)\n",
+        "                \n",
+        "                self.hiddens = nn.ModuleList ([nn.Linear(n_hidden, n_hidden) for i in range(w)])                            \n",
+        "                for m in self.hiddens:\n",
+        "                    nn.init.kaiming_normal_(m.weight)   \n",
+        "                \n",
+        "                self.predict = torch.nn.Linear(n_hidden, n_output) \n",
+        "                nn.init.kaiming_normal_(self.predict.weight)\n",
+        "        \n",
+        "            def forward(self, x): \n",
+        "                x = self.hidden1(x)\n",
+        "                # x = self.BN(x)\n",
+        "                # x = self.Dropout (x)\n",
+        "                x = F.relu(x)   \n",
+        "                \n",
+        "                for m in self.hiddens:\n",
+        "                    x = m(x)\n",
+        "                    # x = self.BN(x)\n",
+        "                    x = F.relu(x) \n",
+        "                              \n",
+        "                x = self.predict(x)\n",
+        "                # x = self.BN_3(x)\n",
+        "                # x = self.Dropout (x)\n",
+        "                return x\n",
+        "   \n",
+        "        def plotCurve(x_vals, y_vals, \n",
+        "                        x_label, y_label, \n",
+        "                        x2_vals=None, y2_vals=None, \n",
+        "                        legend=None,\n",
+        "                        figsize=(3.5, 2.5)):\n",
+        "            # set figsize\n",
+        "            plt.xlabel(x_label)\n",
+        "            plt.ylabel(y_label)\n",
+        "            plt.plot(x_vals, y_vals)\n",
+        "            if x2_vals and y2_vals:\n",
+        "                plt.plot(x2_vals, y2_vals, linestyle=':')\n",
+        "            \n",
+        "            if legend:\n",
+        "                plt.legend(legend)\n",
+        "        #training \n",
+        "        print (\"\\n=== train begin ===\")\n",
+        "        \n",
+        "        net = Net()\n",
+        "        print(net)\n",
+        "        if torch.cuda.is_available():\n",
+        "            net = net.cuda()    \n",
+        "        train_ls, test_ls = [], []\n",
+        "        loss = MAPELoss() \n",
+        "        n_epochs = 10\n",
+        "        optimizer = optim.Adam(net.parameters(), lr=lr, weight_decay=0.0001)\n",
+        "        for epoch in range(n_epochs):\n",
+        "            for x, y in train_loader:\n",
+        "                ls = loss(net(x).view(-1, 1), y.view(-1, 1))\n",
+        "                optimizer.zero_grad()\n",
+        "                ls.backward()\n",
+        "                optimizer.step()\n",
+        "            train_ls.append(loss(net(train_features).view(-1, 1), train_labels.view(-1, 1)).item())\n",
+        "            test_ls.append(loss(net(test_features).view(-1, 1), test_labels.view(-1, 1)).item())\n",
+        "            if epoch % 100 == 0:\n",
+        "                print (\"epoch %d: train loss %f, test loss %f\" % (epoch, train_ls[-1], test_ls[-1]))\n",
+        "        print (\"plot curves\")\n",
+        "        plotCurve(range(1, n_epochs + 1), train_ls,\"epoch\", \"loss\",range(1, n_epochs + 1), test_ls,[\"train\", \"test\"])\n",
+        "        plt.text(60, 0.7, 'Loss=%.4f' % test_ls[-1], fontdict={'size': 20, 'color':  'red'})\n",
+        "        fig_name_1 = '{}-{}_1.png'.format(i,j)\n",
+        "        plt.savefig(fig_name_1, format='png', dpi=300)            \n",
+        "                   \n",
+        "        #plotting\n",
+        "        net.eval()\n",
+        "        predict=net(test_features)\n",
+        "        predict=predict.cpu()\n",
+        "        predict=predict.data.numpy()  \n",
+        "        plt.figure()\n",
+        "        sns.regplot(x=predict, y=test_labels.cpu().data.numpy(), color='g') \n",
+        "        fig_name_2 = 'NN_rank_{}-seed_{}.png'.format(i,j)\n",
+        "        plt.savefig(fig_name_2, format='png', dpi=300)\n",
+        "         \n",
+        "        #save the models\n",
+        "        net_name = 'NN_rank_{}-seed_{}.pt'.format(i,j)\n",
+        "        torch.save(net.state_dict(), net_name)\n",
+        "        \n",
+        "endtime = datetime.datetime.now()\n",
+        "Rtime = endtime - starttime\n",
+        "print(Rtime)"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "4\n",
+            "4\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=110, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (1): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (2): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (3): Linear(in_features=110, out_features=110, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=110, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 0.941584, test loss 0.901918\n",
+            "plot curves\n",
+            "4\n",
+            "4\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=110, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (1): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (2): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (3): Linear(in_features=110, out_features=110, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=110, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 0.930210, test loss 0.918476\n",
+            "plot curves\n",
+            "7\n",
+            "7\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=229, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (1): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (2): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (3): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (4): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (5): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (6): Linear(in_features=229, out_features=229, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=229, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 1.071067, test loss 0.963045\n",
+            "plot curves\n",
+            "7\n",
+            "7\n",
+            "\n",
+            "=== train begin ===\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=229, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (1): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (2): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (3): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (4): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (5): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (6): Linear(in_features=229, out_features=229, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=229, out_features=1, bias=True)\n",
+            ")\n",
+            "epoch 0: train loss 1.008250, test loss 1.001437\n",
+            "plot curves\n",
+            "0:00:06.980756\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "iVBORw0KGgoAAAANSUhEUgAACJcAAAEMCAYAAABzisUIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdZ3hUZcLG8Xtm0kjvjRAChBIgAtKkKCUiqCAIFgQFRXHtr4gioEtTVFwsq4AornRFEBHIIqiggEoXIUiVHkgoCSGQAElm5v1wlhITJEAyk/L/Xddck3PmOc+5z7DmS+59HpPdbrcLAAAAAAAAAAAAAAAAKITZ2QEAAAAAAAAAAAAAAABQelEuAQAAAAAAAAAAAAAAwGVRLgEAAAAAAAAAAAAAAMBlUS4BAAAAAAAAAAAAAADAZVEuAQAAAAAAAAAAAAAAwGW5ODtAcTl79qy2bNmikJAQWSwWZ8cBgDLBarXq2LFjql+/vjw8PJwdBwAAAAAAAAAAAEApVG7KJVu2bFHv3r2dHQMAyqSZM2eqSZMmzo4BAAAAAAAAAAAAoBQqN+WSkJAQScYfSMPDw52cBgDKhtTUVPXu3fvC71AAAAAAAAAAAAAA+KtyUy45vxVOeHi4oqKinJwGAMoWthMDAAAAAAAAAAAAcDlmZwcAAAAAAAAAAAAAAABA6UW5BAAAAAAAAAAAAAAAAJdVbrbFAYC/Y7PZlJycrKysLGdHcQovLy9FRUXJbKZTCAAAAAAAAAAAAODqUC4BUCEcP35cJpNJtWvXrnAFC5vNpkOHDun48eMKDQ11dhwAAAAAAAAAAAAAZUzF+gsrgAorIyNDYWFhFa5YIklms1lhYWE6efKks6MAAAAAAAAAAAAAKIMq3l9ZAVRIVqtVrq6uzo7hNK6ursrLy3N2DAAAAAAAAAAAAABlEOUSSb9tP6pHRi3RjG+36eTpc86OA6CEmEwmZ0dwmor87AAAAAAAAAAAAACuD+USSbFV/FWraoC+/GGnHhv9vf6zYIvSM886OxaAcuzDDz9UTk7OVV+XlJSkgQMHlkAiAAAAAAAAAAAAACgc5RJJvl5uGtK3mca/1E43xUdowYrdemz09/po7iYdTc92djwA5dC4ceOUm5tb4PyVtq6Jj4/XO++8U1KxAAAAAAAAAAAAAKAAF2cHKE2iw301sFdj9bqtjub+uEvfrdmvJav3q13jKronoaYqh3g7OyKAcmDkyJGSpJ49e8psNqty5coKCAjQ3r17lZWVpfnz52vgwIHau3evcnNzFR0drTfeeEN+fn5as2aNxowZo6+//lrJycnq0aOHevbsqeXLl+vMmTMaPXq0mjRp4uQnBAAAAAAAAAAAAFCeUC4pRESwl565t6Huv7W25i3/U0tW7dOy9QfUukFl3XtrLcVE+Do7IoDrsGz9AX2/9kCJzN2hWbTaN4n+2zHDhw/X559/rlmzZsnLy0uDBw/Wtm3bNGPGDHl6ekqSXnnlFQUGBkqS3nvvPU2aNEkvvvhigbkyMjLUsGFDDRgwQAsWLNDYsWM1a9as4n8wAAAAAAAAAAAAABUW5ZK/ERJQSY93i9e9CTU1f/luLfp1r1b8fkjN64XrvltrqVZ0gLMjAignOnXqdKFYIknz58/XwoULlZubq+zsbMXExBR6naenp9q1aydJatiwocaMGeOIuAAAAAAAAAAAAAAqEMolRRDg46GHO9dTj/Y1lbhyjxas3KOB/16hRrVCdN+ttVS/RrCzIwK4Cu2bXHl1EUe7tFiyfv16ffHFF5o1a5YCAwO1cOFCzZ49u9Dr3NzcLvxsNpuVl5dX4lkBAAAAAAAAAAAAVCyUS66Cj6ebHuhYR13b1NDiVfs076fdGjLhF9WtFqj7b62tRrVDZDKZnB0TQBng5eWl06dPy8vLq8BnmZmZ8vb2lr+/v3JycjR37lwnJAQAAAAAAAAAAAAAA+WSa+Dp4aru7WrqztbV9f2a/Zq7bJeGT1ql2Cg/3XdrbTWvFy6zmZIJgMvr16+f+vTpIw8PD1WuXDnfZzfffLMWLFigjh07KiAgQE2aNFFSUpKTkgIAAAAAAAAAAACo6Ex2u93u7BDFITk5WQkJCVq6dKmioqIceu/cPJt+3HBQXy3dpZS0LEWH++i+hFpq3bCyLJRMgFJh27ZtiouLc3YMpyrsO3Dm704AAAAAAAAAAAAAZYPZ2QHKA1cXs25rXlUfvdxeA3s3liSNnblBT45Zqu/W7Fduns3JCQEAAAAAAAAAAAAAAK4N2+IUI4vFrLY3RumWhpW15o8Uzf5hpz6c/bu++G6HerSLVYfmVeXuanF2TAAAAAAAAAAAAAAAgCKjXFICzGaTWsRH6qb6Efptx1HN/mGnPp6XpC9/2Km729RQpxYx8vRwdXZMAAAAAAAAAAAAAACAK6JcUoJMJpMa1wlT4zph2rL7uL78YacmJ27VnKW7dNctNdSldTV5e7o5OyYAAAAAAAAAAAAAAMBlOaxcsnfvXg0ePFgZGRny9/fXmDFjFBMTk2/MsWPHNGzYMCUnJysvL09PPPGEunbt6qiIJap+jWDVrxGsnQdOaPYPO/X5ku2a99OfurNVNXW9pYb8fdydHREAAAAAAAAAAAAAAKAAs6NuNHz4cPXq1UtLlixRr169NGzYsAJj3nrrLdWvX18LFy7UzJkz9d577yklJcUh+ex2u0PuUys6QK/2a64PBrZVk7gwzf1xlx4d/b0mfZOk4xlnHJIBAAAAAAAAAAAAAACgqBxSLklLS9PWrVvVuXNnSVLnzp21detWpaen5xu3fft23XzzzZKkwMBA1alTR99++22B+TIzM5WcnJzvlZqael0Zjy+aqPSfPr+uOa5GtUg/DXqoiT56OUG3NKys//6yV/3fMEomuXlWh+UAAAAAAAAAAAAAAAD4Ow4pl6SkpCgsLEwWi0WSZLFYFBoaWmBVknr16mnRokWy2+06ePCgNm7cqMOHDxeYb+rUqUpISMj36t279zXns9tt53+4eM7mmIJH5RBv/V/PRvpkyK1KaBqtBSv3aMiEX5R2klVMgPLsww8/VE5OjtOuBwAAAAAAAAAAAICicti2OEUxePBgHT9+XF27dtXo0aPVokWLC4WUS/Xt21dLly7N95o5c+Y139dkMivkzicV0LaXJOlcym4lf/x/Ondk3zXPebVCAz31zL0NNbhPU+1PydQL7y/X9n3pV74QQJk0btw45ebmOu16AAAAAAAAAAAAACgqF0fcJCIiQkeOHJHVapXFYpHVatXRo0cVERGRb1xgYKDGjh174bh///6KjY0tMJ+vr698fX2LPafJZJIk2a15svgEydUvRJJky8uR2cWt2O9XmFYNIlU51FujJ6/RkAk/64nuDdTxpqoOuTdQkRyePkw+N7SVT4P2slvzlPL5KPk0TJBPfBvZcs8pddZo+TbuKO+6rWQ7m6XUOWPk1/QOedW5SdbsTB2ZO1Z+zbvIq1ZT5Z0+oaPz3pN/y7vlWaPRFe89cuRISVLPnj1lNpv10Ucfafz48dqxY4fOnTun5s2ba8iQIbJYLBo3bpwSExPl7u4uk8mkadOm6b333st3/fTp00vkdyIAAAAAAAAAAAAASA5auSQoKEhxcXFKTEyUJCUmJiouLk6BgYH5xp04cUJ5eXmSpFWrVmnnzp3q3LmzIyLm4xFVW5EPjpTZw0t2u12pX7ym44snOez+MRG+evf5NqpfI1jj5vyuj+ZuUm6ezWH3B1Cyhg8fLkmaNWuW5s+fr/Hjx6tp06b66quvNH/+fKWnp2vu3LnKyMjQlClT9M0332j+/PmaMWOGPD09C1xPsQQAAAAAAAAAAABASXLIyiWSNGLECA0ePFgTJkyQr6+vxowZI8lYneS5555TfHy8Nm/erNGjR8tsNisgIEATJ05UpUqVHBWxcHabKlWNl4tfsHFot8t2LlsWD68Sva2Pp5tGPHaTpi3apq9/+lP7UjI1uG9TBfh4lOh9gYoi8qFRF342WVzyHZtd3fMfe3jlO7Z4+uY7dvEOyHd8tZYtW6bNmzdr8uTJkqSzZ88qLCxMPj4+io6O1qBBg9S6dWu1bdtW3t7e13wfAAAAAAAAAAAAALgWDiuX1KhRQ3PmzClwftKkiyuCtGnTRm3atHFUpCIxmS0KuOW+C8fZO9bo2H8/UuRDo+QWWrLb1VgsZj3SpZ5qRPnp31/+rgHvLdfQh5upVnRAid4XgGPZ7XZNmDBBVapUKfDZ7Nmz9dtvv2n16tXq3r27Pv30U9WpU8cJKQEAAAAAAAAAAABUVA7ZFqc8cQ2KlFfdlnINjpIk5WWmyW63l+g9b2kUpX89e7MsZpMGj/9Zy9YfKNH7ASh5Xl5eOn36tCSpffv2+uSTT2S1WiVJ6enpOnjwoE6fPq309HQ1a9ZMzz33nGrVqqVdu3YVuB4AAAAAAAAAAAAASpLDVi4pL9xCohVy+z8kSXZrrg5P/6c8ousqtMszJXrf6pX99O7zbfT29PV674uN2p18Uo90qScXC/0goCzq16+f+vTpIw8PD02cOFETJ05U165dZTKZ5OrqqqFDh8rV1VXPPvuszp49K7vdrrp16+q2224rcP306dPl6+vr5CcCAAAAAAAAAAAAUF6Z7CW97IaDJCcnKyEhQUuXLlVUVJRD7mm3WXU6ablc/EJUKSZe9rxc5WUel2tgRInd02q16bPEP7RgxR7dEBusQQ81kZ+3e4ndDygvtm3bpri4OGfHcKrCvgNn/O4EAAAAAAAAAAAAULaw7MV1MJkt8mnQXpVi4iVJJzcs1sGPn1dO2uESu6fFYlb/rvEa8EAjbduXrhfeX67dyRkldj8AAAAAAAAAAAAAAFCxUS4pRt71Wiso4SG5BUVKks4e/lO2vJwSuVf7JtEa80xr2Wx2DRr3s5b/llwi9wEAAAAAAAAAAAAAABUb5ZJi5OIdIL9mnSVJtnPZSv18pNIWTyqx+9WsEqB3B7RRzSr+Gjtzgz5b+IesVluJ3Q8o68rJLmDXpCI/OwAAAAAAAAAAAIDrQ7mkhJjdPRXa40X53dRVkmTNOqkzB7YW+30CfDz02j9a6s5W1TTvpz814tPVOpVdMqulAGWZxWJRbm6us2M4TW5urlxcXJwdAwAAAAAAAAAAAEAZRLmkBHlWayC34ChJ0sm1C5UyY7jyMtOK/T6uLmY90f0GPXtfQ23ZnaYX3l+ufSmZxX4foCzz9/fXkSNHZLNVvNV9bDabjhw5Ij8/P2dHAQAAAAAAAAAAAFAG8X9jdxD/1vfKo0pdufgGSZJOb/1FlarWl8Wr+P7Ye1vzqooO99GbU9bqxQ9WaEDPG9WqQWSxzQ+UZcHBwUpOTtaOHTucHcUpvLy8FBwc7OwYAAAAAAAAAAAAAMogyiUOYnZ1l2fsjZKMLXKOLfhQvo07KqjDI8V6nzpVA/XegLZ6c8pavTVtne5NqKneneJkMZuK9T5AWWM2mxUdHe3sGAAAAAAAAAAAAABQ5rAtjhNYvPxU+bGx8m/VQ5KUczxZp7askN1ePNt1BPp66I2nWqnjTVU1Z+kuvfaf1Tp9JrdY5gYAAAAAAAAAAAAAABUL5RIncQuOksXTV5KU+dt3Sls8SbYzWcU2v6uLRc/c21BP3dNAv+88poHvL9eB1Mximx8AAAAAAAAAAAAAAFQMlEtKgaAODyuy7xuyePpIkk6snK2cYweKZe7bW8Ro9JOtlH0uTy9+sEKrklKKZV4AAAAAAAAAAAAAAFAxUC4pBUwms9xCqkiS8jLTlLFmobL3/F5s89erHqT3nm+jqFAfvTFlrT5fsl02m73Y5gcAAAAAAAAAAAAAAOUX5ZJSxsU3SNFPTZBv406SpOw9m5T+0+ey5Z67rnmD/Svpradbq32TKvriux16Y8paZZ/NLY7IAAAAAAAAAAAAAACgHKNcUgpZPH1kdnGTJJ09uFWnt/4ik/n6/6ncXC16vmcjPd4tXuu2HdHAf6/QoWOnr3teAAAAAAAAAAAAAABQflEuKeUC2zygqEfHymRxld1mVerst5S9e+M1z2cymdTl5up6/R8tlZmVoxfeX651W1OLMTEAAAAAAAAAAAAAAChPKJeUAWb3SpIk6+kTys1Ilf1/W+TY7fZrnjM+NljvPd9G4UFeeu2zNfryhx3XNR8AAAAAAAAAAAAAACifKJeUIS6+wYp67B151m4uSTr123dKnfOWbDlnrmm+0EBPjXmmtdo0itKMb7frrWnrdOZcXnFGBgAAAAAAAAAAAAAAZRzlkjLGZLbIZDJJ+t/KJTabTK4exrH16oshHm4ueqHXjXr0rnpanZSilz5YoZTjWcWaGQAAAAAAAAAAAAAAlF2US8owvyadFHbfEJlMJtnOZevgR8/qVNLyq57HZDKpW5tYjXy8hdIzz2rA+8u1adexEkgMAAAAAAAAAAAAAADKGsolZdz5VUxsuefkHhkrt6DKF46vdiWThrVC9e7zbRTk56E3pqzV4WOniz0vAAAAAAAAAAAAAAAoWyiXlBMu3gEK6z5Q7pGxkqQTK2credILsuWeu6p5woO8NPzRm2Qxm/XGlLU6e+7qt9oBAAAAAAAAAAAAAADlB+WScqpSlbryimshs6u7JCkv83iRrw0N9NRLDzbWgSOn9OGc32W320sqJgAAAAAAAAAAAAAAKOUol5RTnjUbK7DNA5Kk3IyjOjDhaZ1c/22Rr29UO1QPdorTio2HtPDnPSUVEwAAAAAAAAAAAAAAlHKUSyoAi6evAlrfK69azSRJeZlpyjudccXr7mlfU83rheuzBX/ojz1pJR0TAAAAAAAAAAAAAACUQpRLKgCzm4cCWt8jF98gSVLa0qk69OlA2fJy/v46s0kDHrhRoYGeGjNtndIzzzoiLgAAAAAAAAAAAAAAKEUol1RAgW16KqjjozK7uEmSsndvlN1mLXSsVyVXDX24mbLP5WnMtHXKs9ocGRUAAAAAAAAAAAAAADgZ5ZIKyDUwUt5xLSVJ51L3KnXW68rcsPiy42MifPXsvQ21dW+6Jif+4aiYAAAAAAAAAAAAAACgFHBxdgA4l1tYjMLuHaxKMfGSpHMpu2W3WeVRuVa+cW1ujNKOAye0YMUe1Y4O0C2NopwRFwAAAAAAAAAAAAAAOBgrl1RwJpNJXrWayuzmIUk6sXK2jswdK7s1t8DYfl3qKS4mUB/M/l37UzIdHRUAAAAAAAAAAAAAADiBw8ole/fu1f3336+OHTvq/vvv1759+wqMSUtL0+OPP64uXbro9ttv14gRI5SXl+eoiJAU2u15hd83WCaLq+x2u2x5ORc+c7GYNbhvU3m6u+iNKWuVdaZgAQUAAAAAAAAAAAAAAJQvDiuXDB8+XL169dKSJUvUq1cvDRs2rMCYiRMnqkaNGlq4cKEWLFigP/74Q999952jIkKS2a2S3MOrS5JOrl2ow1NfkfXM6QufB/p66OU+TZWanq33vvhNNpvdWVEBAAAAAAAAAAAAAIADOKRckpaWpq1bt6pz586SpM6dO2vr1q1KT0/PN85kMikrK0s2m005OTnKzc1VWFiYIyKiEG5BleUWWlVmD8985+tVD1K/LvW05o9Uzf1xl5PSAQAAAAAAAAAAAAAAR3BxxE1SUlIUFhYmi8UiSbJYLAoNDVVKSooCAwMvjHvqqaf07LPPqnXr1jpz5ox69+6txo0bF5gvMzNTmZmZ+c6lpqaW7ENUQJ6xjeUZa3z/1jOnlJueKo/KNSVJd91cXTv3n9CMb7epZhV/NawV6syoAAAAAAAAAAAAAACghDikXFJUixcvVu3atTV16lRlZWWpf//+Wrx4sTp16pRv3NSpUzVu3DgnpayY0r6fouxd61Tl6Y9k8fCSyWTSM/c11L7UTL09fYPef6GNQgM8rzwRAAAAAAAAAAAAAAAoUxyyLU5ERISOHDkiq9UqSbJarTp69KgiIiLyjZsxY4buuusumc1m+fj4qH379lqzZk2B+fr27aulS5fme82cOdMRj1JhBd3aV6F3vyCLh9eFc5XcXTT04WbKs9r05tR1ysm1OjEhAAAAAAAAAAAAAAAoCQ4plwQFBSkuLk6JiYmSpMTERMXFxeXbEkeSoqKitGLFCklSTk6OVq1apZo1axaYz9fXV1FRUfle4eHhJf8gFZjF01ee1RtKkrJ3b1T6jzNlt9tUOcRbAx64UX8ezNAn3yQ5OSUAAAAAAAAAAAAAAChuDimXSNKIESM0Y8YMdezYUTNmzNDIkSMlSf3791dSklFKGDp0qDZs2KAuXbqoW7duiomJ0X333eeoiCiiM3s3K/vP32TPzZEktYiP0L0JNbVk9X59v2a/k9MBAAAAAAAAAAAAAIDi5OKoG9WoUUNz5swpcH7SpEkXfo6OjtbkyZMdFQnXKDChj+w5Z2V285DdZpU9N0e9O8Vp14EMffT1ZlWL9FNsFX9nxwQAAAAAAAAAAAAAAMXAYSuXoPwwmUwyu1eSJKUt+Y8OT3tFJmuOXnywsfy83fXm1LXKzMpxckoAAAAAAAAAAAAAAFAcKJfgunjWbi6vuJYyu7rLz9tdQ/o2VXrmOb0zc4OsNruz4wEAAAAAAAAAAAAAgOtEuQTXxbN6AwW0vkeSlJN2WFV0WE90j9dvO47qiyXbnZwOAAAAAAAAAAAAAABcL8olKDZp30/W0W/+rQ6NI9WhWbS+/GGn1v6R6uxYAAAAAAAAAAAAAADgOlAuQbEJ7fqcwu8fKrOrm/7R/QbViPLTu59v0OHjp50dDQAAAAAAAAAAAAAAXCPKJSg2lko+cg+LkSSd3fS9XojdJYtZenPKOp3NyXNuOAAAAAAAAAAAAAAAcE0ol6BE5J1IkduZ4xr4wI3an5qp8XM2yW63OzsWAAAAAAAAAAAAAAC4Si7ODoDyKfDWhyWbVeEWF/VJiNHspTv036oB6ty6urOjAQAAAAAAAAAAAACAq0C5BCXCZDJJFhfZ7Xa1yliomJBDemO+m2pU9ldctUBnxwMAAAAAAAAAAAAAAEXEtjgoUSaTSf4tuiqmw30KDvDWW9PW6UTmWWfHAgAAAAAAAAAAAAAARUS5BCWuUky8gm9spyEPN1VozgFN/+xr5Vltzo4FAAAAAAAAAAAAAACKgHIJHCYmwlePVN6phieXaWriFmfHAQAAAAAAAAAAAAAARUC5BA5jMplUu98I7anzsL5ZsVcrNx6U3W53diwAAAAAAAAAAAAAAPA3KJfAoSweXurdo6XiqgZo3zcfa9+8CRRMAAAAAAAAAAAAAAAoxSiXwOFcXcwa1KeJZHHV2u1pOnMuz9mRAAAAAAAAAAAAAADAZVAugVME+3uqYe+n9fmJBnp/1kblZByVNfuUs2MBAAAAAAAAAAAAAIC/oFwCp4mvEaxHOtfT6qTD2jl5lFJnv8kWOQAAAAAAAAAAAAAAlDKUS+BUXW+poVYNKuvT1Bt0olYXmUwmZ0cCAAAAAAAAAAAAAACXoFwCpzKZTHru/kbKCaqht5Zk6NiJMzq1+Sdl797o7GgAAAAAAAAAAAAAAECUS1AKVHJ30ZC+zZSTa9OYqauVsW6RTq5bxBY5AAAAAAAAAAAAAACUApRLUCpUCfPR8z0bacfBTC3yvkdhdw9gixwAAAAAAAAAAAAAAEoByiUoNVreEKke7WKVuDZFyzYdlzXrpI4uHKe8k8ecHQ0AAAAAAAAAAAAAyj+TyXhVNFOnSs2aSd7ekp+f1LatlJh4fXPa7VKHDhe/07y8wsf9+KN0xx1SUJDk7i7FxkqDB0unTl1+7q1bpfvuk0JDJQ8PqXZtafhw6cyZgmP37buYobBXz55FehyXIo0CHOSh2+O062CGPpq7SdV842TZuU6eNRrJ2y/E2dEAAAAAAAAAAAAAAOXNiy9K77wjRUVJ/ftLOTnSrFlSly7Shx9KzzxzbfOOG2cURzw8pLNnCx8zcaL01FOSi4vUvbuRYcMGacwYadEiaeVKo+xyqTVrpPbtpdxc6Z57pCpVpGXLpFGjpKVLjZe7e8F7NWggdetW8Hz9+kV6HJPdbrcXaWQpl5ycrISEBC1dulRRUVHOjoPrkHHqnAa895MsFrPeebqZ/Pz9rnwRgGvC704AAAAAAAAAAABccH7VkvJRI7iyX3+VWrWSatSQ1q2TAgKM8/v2SY0bS1lZ0vbtUkzM1c27Y4fUqJH03HNGUWX/fqMM4nLJ+h8pKVL16saKJr/8Yqycct6bb0pDh0rPPit98MHF81arFB8vbdsmzZ8v3XWXcd5mM1YymTvXuHbw4IvX7NsnVasm9e0rTZlydc9xCbbFQanj7+OuwX2bKu3kWY2d9YesVpvOJu+QNeuks6MBAAAAAAAAAAAAACTp3DnprbeMsoOnp+TrK918szR7duHjFyyQEhKkiAhjZY3ISKlNG2nChPzj9uyRHn/c2B6mUiUpMNC4xxNPSGlpxfsMEyca76+8crFYIhllkqefNp5x8uSrmzMvT3roIaM4MnLk5cd9+62xokm3bvmLJZI0aJDx3J99JmVnXzy/fLlRLLnllovFEkkym6W33774TCVQDqJcglKpdtVAPdXjBv2+65imzVurlJkjdGLlZX4JAQAAAAAAAAAAAAAcJydH6thRGjLEKFM8/bRRqNi5U7r/fmPVjUt98onUtau0daux3czAgdIdd0hnzuQvb6SkSE2bGufq1TNW/njoIWPljenTjc+L07JlxnunTgU/u/32/GOK6vXXpY0bjVVCCtue5rzUVOO9evWCn1ksUtWqxsopa9YULW/16lKtWsYqKXv2FPz88GHp44+lN94w3jdvLvIjSZLLlYcAztGheVXtTz2lr1fsVmz7vmrVro2zIwEAAAAAAAAAAAAA3nnHWEXj9tuNFUnOb/cyfLixCsebb0qdO0stWxrnP/5YcnOTNm2SQkPzz3X8+MWfv/pKSk+X3n9f+r//yz8uK8tYoeO8jAxj3NXo1k1q2PDifIcOSd7exmoqf1WzpvG+c2fR51+3Tho92tiWpkmTvx8bHGy8791b8DObzSiJSMYWO+3aXfxZMkokhalZ08i7c6ex1c+lvv/eeF2qbVtp6lQpOvrvs4pyCUq5RzrX1cGjp1VA2mYAACAASURBVPTOT8fkXydL9au7y56XK7Pr3zS8AAAAAAAAAAAAAAAl57PPJJNJevfdi8USySiO/POf0mOPSZ9+erFcIhnjXF0LznW+ZHGpSpUKnvPyyn+ckfH3284UJibmYrnk5Enj3c+v8LHnz2dkFG3uM2eMVVbq1ZOGDbvy+I4dje/km2+k9evzl1HGjjVKNpJ04sTF89eS2dPT+Dfp1u3iKimbN0sjRkg//mhsVfT77wW/379gWxyUahaLWS892EThQV4aM2WN9k/+p9K+v8o9rQAAAAAAAAAAAAAAxePUKenPP6XISKlOnYKft29vvG/cePFc795SdrZUt640YIBRqDh2rOC1d91lrCTy9NNSjx7Gdjp//CHZ7QXHxsQY56/m9fDDxfENFG7QIGM7mqlTCy/R/FXVqsZKL7m5UqtWUq9e0ksvGWWPwYOlG24wxpmvs9YRGiqNGiXdeKPk72+8brlF+u47qXlz49/y00+vOE2RU6xevVoHDx6UJB09elQvv/yyhgwZomOF/YMDxci7kquGPdpcVrtJK4/6yRxW09mRAAAAAAAAAAAAAKBiOr96RmFbyVx6/tLVM154wShdVK0qffCBdPfdUliYsd3L+vUXx1WtKq1dK3XvLv3wg/SPf0j161+8rjidX+Xj/PP81fnz/v5Xnmv5cmn8eOnVV6UGDYqe4dVXja2AmjeXFi405sjOlhITpZtvNsZcuo1QcWZ2cTFWmJGkFSuuPPzKMxpGjhyp//znP5KkMWPGSJLc3d31z3/+UxMnTizqNMA1iQzx1st9mmj4pDzt3+StoY3sMptNzo4FAAAAAAAAAAAAABXL+YJDamrhn6ek5B93Xp8+xisjQ/r1V2nePGN7nY4dpe3bpZAQY1xcnPTll1JenrRpk1Ey+fBD6f/+z9i65dFHjXEZGdL7719d9m7dLm6L4+UlVa4sHTpkZP5rWWbXLuO9Vq0rz7txo7EyyvDhxqsw51cz2bjxYgbJWKGlR4+C4996y3hv2vTiudq1jfedOwu/x9Vkli5+51lZVxxa5HLJkSNHFBkZqby8PP38889atmyZXF1ddfP5tswV7N27V4MHD1ZGRob8/f01ZswYxcTE5BszaNAg7dix48Lxjh07NH78eCUkJBQ1JsqxhrVC1b9rfX08L0nffjFHLWMsCmhVyH9kAAAAAAAAAAAAAICS4eMj1ahhbAGza5dU8y87T/z4o/F+442FX+/vL91xh/Gy2YyCyYoVBQsWLi5S48bGq2VLYyuXb77JXy4ZOfLqssfE5C92tG8vTZ8uLV4sPfJI/rHffntxzJXUr38x1199+aV0+rTUr59kMklBQVeeb/du6ZdfpPh4Y+5L844ebeQdMiT/NXv2GKWTqlWl6tWvfA9JWr3aeC/C+CKXS7y9vXX8+HHt2rVLNWrUkJeXl3JycpSXl1ek64cPH65evXqpa9eumj9/voYNG6Zp06blG/P2229f+Hn79u3q27dvkcsrqBjubFVN+1Iylb75S6WeOCv/m7rKZCny/4wBAAAAAAAAAAAAANerXz/plVekl16S5s6VLBbj/PHj0muvXRxz3o8/Sm3bGuWKSx09arx7ehrvGzZIsbEFVz05ciT/OMkoitjt1/ccTzxhlEtGjzZWNQkIMM7v22dsUePuXrB0cvy48QoONl6SdOutxqswP/xglEs+/tgozFwqM1Py9c1/Li1N6t3bKN78b1eZC9q0MVZ2WbFCWrBAuusu47zNJr388sVnuvR7/u03o1BjNuefa+lS6b33jJ8ffLDw7Jco8l/lH3zwQd1zzz3Kzc3V0KFD/5fhN1UvQoMlLS1NW7du1eTJkyVJnTt31muvvab09HQFBgYWes1XX32lLl26yM3NragRUQGYTCb94+4bNPLoCc0/kKk3Dp1SregAZ8cCAAAAAAAAAAAAgPLj4Ycv/9mECdKLLxore8yfLzVoYKxCkp0tzZljFEYGDZJat754zd13S97e0k03XSyFrFwprVtnrExyvpgxfbpRwmjd2lgdJSDAWMVj4UKj6PH888X7nC1bSi+8IL37rnTDDdI990g5OcZqI+npxnY8f9mRRePGGSumDB8ujRhxffcfNcpYhaRFCyk01NiiZ8ECY1WWd96Rbr89/3iLRZo82VjB5J57jFd0tFEUWb9eatVKGjAg/zUvvGCsMNOypRQVZZzbvFlatsz4+bXXjM+uoMjlkscff1wdOnSQxWJRdHS0JCksLEyvv/76Fa9NSUlRWFiYLP9rK1ksFoWGhiolJaXQcklOTo4WLlyoKVOmFDpfZmamMjMz851Lvdx+Tih3XF3MeqlvSw389wqN+ewXje4RpvD4ple+EAAAAAAAAAAAAABwZVOnXv6z9983VhD5/nujlPH550YJw8XFKJq8/770wAP5r3nrLWnJEmMVjUWLJA8PY/uWMWOkJ5+UXF2NcQ88IJ07J/36q7GKyZkzUuXKUs+e0sCB+beIKS7vvGNsPzN+vPTJJ8YKHzfeaKzK0rlz8d/vUu3aGd/J/PlGoSQwUEpIMJ71ppsKv6Z5c6OUM3y49N130qlTxnc5bJg0eLBRwrnUQw9J8+YZ13z7rZSbK4WFSffdJz3zjFTE3WRMdvu1rROzevVqmc1mNWvW7Ipjt2zZopdffln//e9/L5y744479K9//Uv16tUrMH7RokWaNGmS5s2bV+h8H374ocaNG1foZ0uXLlXU+bYNyrX9KZn66eOxaum6TZFPTpBXYIizIwFlTnJyshISEvjdCQAAAAAAAAAAAOCyrmpbnAEDBqhx48b65JNPNGXKFFksFvXu3VtPPPHE314bERGhI0eOyGq1ymKxyGq16ujRo4qIiCh0/Ny5c9WjR4/Lzte3b1/dfffd+c6lpqaqd+/eRX0clANVI3xV966HNHH2UoX9d79eejBYpr/u0QUAAAAAAAAAAAAAAK6LuagDd+3apYYNG0qS5syZo2nTpmn27NmaNWvWFa8NCgpSXFycEhMTJUmJiYmKi4srdEuc1NRUbdiwQV26dLnsfL6+voqKisr3Cg8PL+qjoBxpemNNtbgtQSt/P6SvlmxxdhwAAAAAAAAAAAAAAMqdIpdLbDabTCaTDhw4ILvdrtjYWEVEROjkyZNFun7EiBGaMWOGOnbsqBkzZmjkyJGSpP79+yspKenCuHnz5qldu3by8/O7ykdBRdWjXaweqJOl2uve1NrVm50dBwAAAAAAAAAAAACAcqXI2+I0btxYo0aN0rFjx9ShQwdJ0oEDBxQQEFCk62vUqKE5c+YUOD9p0qR8x08++WRRIwGSJJPJpG7dE7Ti0yR9uWC7QqKrqlok5SQAAAAAAAAAAAAAAIpDkVcuefPNN+Xr66vatWvrmWeekSTt2bNHffr0KbFwQFF5BoWp2ZPDZffw02ufrVHGqXPOjgQAAAAAAAAAAAAAQLlQ5JVLAgIC9MILL+Q717Zt2+LOA1yzQF8PvdozTutnTtD7n1n1ytO3ydXF4uxYAAAAAAAAAAAAAACUaUVeuSQ3N1cffPCBEhISFB8fr4SEBH3wwQfKyckpyXzAVYkOclETz2TlpezShK82y263OzsSAAAAAAAAAAAAAABlWpFXLvnXv/6lzZs3a+TIkYqMjNThw4c1YcIEnT59WkOHDi3JjECRuQVFqvrzn6jOj/v15fc7VTXCV93a1HB2LAAAAAAAAAAAAAAAyqwil0sWL16s+fPnKyAgQJJUvXp11a1bV127dqVcglLF7O6pXrfVUdb+bfoycYOqhHmrcZ0wZ8cCAAAAAAAAAAAAAKBMKvK2OJfbXoRtR1Aa2bIy1OHEl+oetENvT1+vg0dOOTsSAAAAAAAAAAAAAABlUpHLJZ06ddKTTz6plStXavfu3VqxYoWefvppderUqSTzAdfExSdAYT1e0i0PPyU3F4te+2yNTmXnODsWAAAAAAAAAAAAAABlTpHLJS+99JJatGihUaNGqXv37nr99dfVvHlzDRo0qCTzAdfMq2YThYUHaUjfxjpx4pTenrZeVqvN2bEAAAAAAAAAAAAAAChTXP7uw1WrVuU7btasmZo1a5bv3IYNG9SiRYviTwYUA7s1T/4//1tD6wdr2CazPl2wRf+4+wZnxwIAAAAAAAAAAAAAoMz423LJK6+8Uuh5k8kkSbLb7TKZTFq6dGnxJwOKgcniIo/oeooNilA3/2B9s3y3osN9dXuLGGdHAwAAAAAAAAAAAACgTPjbcsmyZcsclQMoMYFtH5AkPVzProNHTunjrzcrKsRb8bHBTk4GAAAAAAAAAAAAAEDpZ3Z2AMBRzmz/VU/VSVFEsJfenLpOqWlZzo4EAAAAAAAAAAAAAECpR7kEFUb2n78pd/d6vfpIE9ntdr322Rpln811diwAAAAAAAAAAAAAAEo1yiWoMII7PabIPq+pcqifBvdpquSjpzV25gZZbXZnRwMAAAAAAAAAAAAAoNSiXIIKw+xWSSazRbbcc6rllqrHu9bXuq1HNH3RVmdHAwAAAAAAAAAAAACg1KJcggonfek0pX75hjo2DNTtLWI098c/9eOGg86OBQAAAAAAAAAAAABAqeTi7ACAo/m3ukdecS3k4u2vx+/2VfLR0/pw9u+KDPZS7aqBzo4HAAAAAAAAAAAAAECpwsolqHBcfAJUqWp9SZLZlqvBfZsqyM9Doyev1fGMM05OBwAAAAAAAAAAAABA6UK5BBXW6e2rdHD8U/K0ZurVfs11Nseq1yev0dmcPGdHAwAAAAAAAAAAAACg1KBcggrLPayaPKrWk0wWVQ331YsPNtaeQyf171kbZbfbnR0PAAAAAAAAAAAAAIBSgXIJKizXgHCF3f2CXHwCJEnN6oar7x119fOmw/ryh51OTgcAAAAAAAAAAAAAQOlAuQQVXt7pDB1LnCBr1kl1bxerdo2jNHPxdv26+bCzowEAAAAAAAAAAAAA4HSUS1Dh2c5k6vS2X3X20E6ZTCY9c29D1a4aoHe/+E17Dp10djwAAAAAAAAAAAAAAJyKcgkqPLeQaEU/+7G8ajU1jl0teuXhZvKp5KrXPlujE6fOOjkhAAAAAAAAAAAAAADOQ7kEkGTx8JIknU3eLtvZLAX4euiVfs2VmZWjN6esU3omBRMAAAAAAAAAAAAAQMVEuQT4n7zM4zo8fZgyVs2TJMVG+WvAA420fX+6Hh61RIM+XKlvlv+p1LQsJycFAAAAAAAAAAAAAMBxXJwdACgtXHyDFdZ9oCpVu+HCudYNKqtquK9+2XxYqzan6D8L/tB/Fvyh6pX91DI+Qi3iI1QlzEcmk8mJyQEAAAAAAAAAAAAAKDmUS4BLeNVuLkmy26yS3SaTxVVVwnzUs0Nt9exQWynHs7QqKUWrkg5rxuLtmrF4uyqHeKvlDUbRJDbKn6IJAAAAAAAAAAAAAKBcoVwC/IU9L1eHZw6XR1QdBSX0yfdZRLCXureLVfd2sUo7eUart6RqVdJhzf3xT81ZukshAZXUIj5CLeMjVScmUBYzRRMAAAAAAAAAAAAAQNlGuQT4C5OLqzyi6sgtLOZvxwX5VdKdrarpzlbVlJmVo7V/pGpVUoq+/XWfFqzYI39vdzWvH66W8ZGKjw2Wq4vZMQ8AAAAAAAAAAAAAAEAxolwCFOKvK5Zcia+Xm25tFq1bm0Ur+2yuNmw7ql+TDmv5b8lasnq/vDxc1LSeUTRpVDtEHm78pwcAAAAAAAAAAAAAKBsc9hfuvXv3avDgwcrIyJC/v7/GjBmjmJiYAuMWLVqkjz76SHa7XSaTSZMnT1ZwcLCjYgL5nN76i6xZJ+XX9I4iX+Pp4aqbG1XWzY0qKyfXqt93HtOvSYe1ZkuqftqQLHc3ixrXCVWL+Eg1jQuTVyXXEnwCAAAAAAAAAAAAAACuj8PKJcOHD1evXr3UtWtXzZ8/X8OGDdO0adPyjUlKStK4ceM0depUhYSE6NSpU3Jzc3NURKCArB1rZD2VLt8mnWQ/d0an/1ipSjVulKt/6IUC1N9xc7WoWb1wNasXrjyrTX/sTtOvSYe1ekuKft2cIheLSQ1qhqhFfKRuqh8uP293Bz0ZAAAAAAAAAAAAAABFY3bETdLS0rR161Z17txZktS5c2dt3bpV6enp+cZNmTJF/fr1U0hIiCTJx8dH7u78sR3OE3LHk4p4cKRMJrNy0g7r+OJJyjm6X5J0LmW39o19SGf2JUmScjOOKOPXr5WXmSZJslvzZLdZL8zlYjGrQa0QPdmjgSb/s6PefuZmdW5dXclHT2vcnN/VZ8RiDZ3wixau3KNjJ844/mEBAAAAAAAAAAAAACiEQ1YuSUlJUVhYmCwWiyTJYrEoNDRUKSkpCgwMvDBu9+7dioqKUu/evZWdna0OHTroySefLLA6RGZmpjIzM/OdS01NLfkHQYVjdq904Wf3iOqKfvYTmT28/veZp7zj28rFzyhD5RzZr/QfZ6pS9UZy8Q1S9q71OvL1O4p6bKzcQqvq7OE/lb1jtfyad5XF00e1IyupdkQ19etST3sPZ+rXpMNalZSiT75J0iffJKlWtL9axEeqZXyEIkO8nfL8AAAAAAAAAAAAAAA4bFucorBardqxY4cmT56snJwcPfbYY4qMjFS3bt3yjZs6darGjRvnpJSoqExmi1x8gy4cuwVFKrjjoxeOvWo3U8xLM2RyMbZycg2MkH/L7nLxDZYk5Rzdr4zVC+XXvKsk6dSmpUr7frKqDpis6pX9FJa1Ux3tu3T2wXu1eutxbdr0pxZ8u1ZT/+ulquG+urlRZXVvW1OuLg5ZcAgAAAAAAAAAAAAAAEkOKpdEREToyJEjslqtslgsslqtOnr0qCIiIvKNi4yMVKdOneTm5iY3NzclJCRo8+bNBcolffv21d13353vXGpqqnr37l3izwL8HbPbxZVO3EKrKjC06oVj34YJ8mnQTpKxEo9HdD0F3tpX5ko+kqTc9BRl79qg6E6P696IQLW3r9LJnG+1peXr+jUpVT9994sO7t6n5/rdKjdXi0OfCwAAAAAAAAAAAABQcTmkXBIUFKS4uDglJiaqa9euSkxMVFxcXL4tcSSpc+fOWr58ubp27aq8vDytXr1aHTt2LDCfr6+vfH19HREdKFYm08VVR9zDq8k9vNqFY/+b7pL/TXddOPaObyP3yFjdVTdWXVpX0473p+tAilmvfeqtV/o1l4d7qVp4CAAAAAAAAAAAAABQTjlsf40RI0ZoxowZ6tixo2bMmKGRI0dKkvr376+kpCRJ0p133qmgoCDdcccd6tatm2JjY3XPPfc4KiJQqriHxci7bitJxpY81R8cKo92j2nz7uMaPmmVss/mOjkhAAAAAAAAAAAAAKAiMNntdruzQxSH5ORkJSQkaOnSpYqKinJ2HKDE/LzpkH6f85lyAmLU/8n75ePp5uxIKMP43QkAAAAAAAAAAADgShy2cgmA4tGybrBuCzuqkJPbNHTCLzpx6qyzIwEAAAAAAAAAAAAAyjHKJUAZY3Z1V2z/t1S/13M6fDxLr45boeMZZ5wdCwAAAAAAAAAAAABQTlEuAcogs4eXGtUJ16iHG+iB3DmaO368UtOynB0LAAAAAAAAAAAAAFAOUS4ByrC6NcMVXrOODp7z05DxP+vQsdPOjgQAAAAAAAAAAAAAKGcolwBlmMniqtgHBurRJ+5TrtWmDybM075DGc6OBQAAAAAAAAAAAAAoRyiXAOVAtUg/vfFgLT1qWaAfPh2nPw9SMAEAAP/f3r2HWVUe5gJ/Z88dEOU+w02RyMU7goBiooJVG1G8nRA1aWK9xFhRc46JaS5ekrQJ1lpz8J60DSY29nhiRYEoidqqxKpJCBHQqAiIyAACCnJnZs4fOcFQOtFEYA3w+z3Pfmbttdfs9X7fM3v/w8v3AQAAAADA9qFcAruJ3gf0Te3xF2RG+WH58h3TM2fe8qIjAQAAAAAAALAbUC6B3UivESfnur86IR3aVubJ792WmbPmFR0JAAAAAAAAgF2ccgnsZrp0qM3Xx/bO8VXP58f33p+fv7Ck6EgAAAAAAAAA7MKUS2A31KXvgHT+9I1p6DQ4f/PPz2T6zEVFRwIAAAAAAABgF6VcArupjj165xuXjMig7mVpmnRtpj/+dNGRAAAAAAAAANgFKZfAbqxdbWU+N/awVFdV5ns/fimP/Of8oiMBAAAAAAAAsItRLoHd3F71vXPo/5yQHgf0yy33zcyPH3m26EgAAAAAAAAA7EKUS2APUFNVkS+fPzT/40Or8qHnbsjU+x8pOhIAAAAAAAAAuwjlEthDVFaU55xPnZmX9jk6d05fm+//+IU0NzcXHQsAAAAAAACAVk65BPYglW3a5uRLP5cThvXJAz+dnanfm5impqaiYwEAAAAAAADQiimXwB6mvFSWvzr7sFw4cHn6L5qcH/zLT9LUZAUTAAAAAAAAAP57yiWwByqVynLSX16Ymf0uzn0z1ufme3+ZxkYrmAAAAAAAAACwLeUS2EOVSuU562Mn5hN/PiDzZ/4yT337q9mwbl3RsQAAAAAAAABoZZRLYA839oT+OeuIvVL5zuL8w/efzsZNjUVHAgAAAAAAAKAVUS4BcuzYj2fVyL/O9N+szte++3TWvv1W0ZEAAAAAAAAAaCWUS4AkycnHHJDPnTMovRb9NL+57X9l9cqVRUcCAAAAAAAAoBVQLgG2GDmkdw4/4YQ8t7ZXrvnezKxeu7HoSAAAAAAAAAAUTLkE2Mqw447J4edckgUNq/O3tz6SZa/MLjoSAAAAAAAAAAVSLgG2MfTAulxzwbCMWPtoFv/rt7LszbeLjgQAAAAAAABAQZRLgP/W4f26ps/Z43L3uuPz13c+m4bla4qOBAAAAAAAAEABlEuAFh10UN9cfPEZWbNuU35w2/ey4ImHio4EAAAAAAAAwE6mXAL8Qf16d8jffPboDMwreeU/Hsm8RW8VHQkAAAAAAACAnUi5BHhP+/fYJ4dccE3ubTopX7jlqdw/5bmsnDU9zc3NRUcDAAAAAAAAYAdTLgHel97dO+Rvxo3KoP5ds+TpKXnzgZvzH0/+Ok1NCiYAAAAAAAAAu7OKogMAu46uHdvkS58emtmv7Jd/m/RYnpw0P//23Mp8ZuCyfGjY0anq1KPoiAAAAAAAAABsZztt5ZJ58+Zl7NixOemkkzJ27NjMnz9/m2smTJiQo446KmPGjMmYMWNy/fXX76x4wB/hoA91zVWfG5vPf2JwNq9bnaZf/CiP3P39vNawquhoAAAAAAAAAGxnO23lkmuvvTbnnntuxowZk0mTJuWaa67J3Xffvc11p59+eq6++uqdFQv4E5VKZfnIoJ4ZfnB9Hn6sZ+5/Yl6+e+PjOfvw6oyqW566j5yVUlVN0TEBAAAAAAAA+IB2ysoly5cvz5w5czJ69OgkyejRozNnzpysWLFiZ9we2IGqKstz2kmD8u2/PiUfHdEnb7348yz/zym5/9HfZP3GzUXHAwAAAAAAAOAD2ikrlyxevDjdunVLeXl5kqS8vDxdu3bN4sWL07Fjx62unTJlSp566ql06dIl48aNy6BBg7Z5v1WrVmXVqq2332hoaNhxAwDe097tqvOZMw7NomP2zw8f/Hn+46fzM/m5xflc7+fTd8SotDvgiKIjAgAAAAAAAPAn2Gnb4rwfH//4x3PJJZeksrIy06dPz6WXXpqpU6emQ4cOW103ceLE3HLLLQWlBP6QHl3a5aoLjsufv7o890x6Nmtfm5N/XVKdIWf0ymH9uhQdDwAAAAAAAIA/0k4pl9TX12fJkiVpbGxMeXl5Ghsbs3Tp0tTX1291XZcu7/7D84gRI1JfX5+XX345Q4cO3eq6T33qUznjjDO2OtfQ0JDzzjtvxw0C+KMctH+nfOOKk/PkLw/M0w+/kAfu/FlO77MqIzssTO8zx6W8dq+iIwIAAAAAAADwPpR2xk06deqUgQMHZvLkyUmSyZMnZ+DAgdtsibNkyZItxy+88EIWLVqUPn36bPN+7du3T8+ePbd61NXV7dhBAH+0Uqksxw7ZN7dd/Wc5f/SBWbl0Wea/Mj93Tn4lK1evT3NzU9ERdwvNTY1FRwAAAAAAAAB2YzttW5zrrrsuX/ziF3Pbbbelffv2GT9+fJLkoosuyuWXX55DDjkkN910U2bPnp1SqZTKysrccMMNW61mAuyaqirLc+bxB+TtI/8q9057MdOeXpAnZ7yWL3X5abp/+NR0GnJi0RF3WUsnfTtl5ZXpMvrSoqMAAAAAAAAAu6mdVi7p27dv7rvvvm3Of+c739ly/LvCCbB72rtddT5z5mEZ/eG++eGDP88rr1XkwakLMrz5tRx3eH1K5aWUlcqLjtmqNTc1Zv1rc1K73yFJkqpu+yVWLgEAAAAAAAB2oJ1WLgH4nR5d2uWqC47L7FcPyZMPzsrN987Ia4/dn2PavJz9L/xWymv3Kjpiq7V65uN5c+rt6f7pb6amR7/sM3xM0ZEAAAAAAACA3ZxyCVCYg/bvlBsv/0ie/NWiTJ+6LP+5tH3+5Z7ZOX/0gemxV3PK2+5ddMTCNW3akLefnZLq+v3TZv/D0+6gESnVtEl1fd+iowEAAAAAAAB7iFLRAYA9W6lUlmOP6Jmrrj4/bUf+ZV6YtzxX//3Defl/X5rF/77tVlp7iubGzUmSsvKKrJ7xk6yb9+skSamqNu0GHm37IAAAAAAAAGCnsXIJ0CpUVZbnzOMPyKgje+f/PPJ8HpvRPy/8ZF2O2fybnDq0W6rLy/aYlUxWPvV/s+aF6elx4d+nrFSenhfemFJN26JjAQAAAAAAAHso5RKgVdm7XXUuOmtIFn1kQN6aMic/+PGLaXz6X3Jk1dz0GXdHKtq0Kzridtfc3Jz1C+ekuvsBKVVUpapzzzTue3CaN21MWVWNYgkAAAAAAABQKOUSoFXq0aVdvvTpoZk1983cP6kpC95sl4bbf56/FeYtFgAADrBJREFUPO2g9K9dnuruB+w2W8NseOPlLP7+Nel8ymfT/vAT0nbA8LQdMLzoWAAAAAAAAABJklLRAQD+kIP7ds5Xrjw9R599Tt5ZtzE33TUtr0/8al58+L6sXb8pTZs3Zt3859O0cX3RUd+35ubmrPrltKya8dMkSXX3A9L19CvT7qAPF5wMAAAAAAAAYFtWLgFavVKpLMce0TNHHVKfyU/OzT3//nZee3xjlj46NYM7rspf5IG81Pe87H3QUdm3/ea0e+vltO13ZMpr9yo6+laaGzenrLwiZWVlWfObZ1NWKqX9oBNSVlamWAIAAAAAAAC0WsolwC6jqrI8Z47slxOH75c581dk3htvZ+Hry3Lf4lPyy19sztrnns2RVXPziXbTc2u7v0jHXn0yoO3KdN+0ID2POyO17Yorm7wz+6ks/8k/p+dnbk557V7pdub/TFlVbWF5AAAAAAAAAN4v5RJgl9OuTVWGHliXoQfWJemf5Jis37A58xtWZd6iQ/Lk/EOz8s2qzPj5a1lT9uvU1c7IJ/+9bbp02SfHtV+QPuVLUhrxqfTp2SEd29ekrKxsh+TcuHRBSjXtUtG+U6q69E5tn0PTvGljUpuUqtvskHsCAAAAAAAAbG/KJcBuoaa6IgP27ZgB+3ZMjt4/SdLU1JwlK47PvNeW5Yyl6zPvjbfzzuJleWfz/PzDPz2XJBnb/pepb7Mx8/uflz7d22e/bm3Tq75DKitKHyhP49rVef2fvpD2g09O5z87P1Vde6frmCs+8DgBAAAAAAAAdjblEmC3VSqVpb5z29R3bpujt5wdlnfWbco333g7895YldKshXl71YpMnT4vGzc3ZdxeD+dXzbV5fK9T06d7+/Tv2Jge+/VKn54ds3e76j94vzUvPZeNS+alw4c/lvI2e6XbmVelpteAHT1MAAAAAAAAgB1KuQTY47SrrczBfTvn4L6dkw9fliQZ09iUN95ck2U/eydNq5vTcUNNZr68NCNLP8icn/XIV9eMSMf2NTm604rs1etD6dm7Pn26752O7SpSW1uTUqks6xe+kLUvP5e9jzo9pYqqtO13ZMEjBQAAAAAAAPjglEsAkpSXl9Kr217pdcZ5SZKTkzQ3NWbZjJqUb6hO7caueX1hQ/7stbsytWFQfvjEIdm3fFku2uvx3LF6VJZXdE376g6prDotNbc8nTbVFamtrkhtzW9//v7z3x5X/vb8/3/996+tKP9gW/IAAAAAAAAAbE/KJQAtKCuVp+vg49M1ySFJmhv3y4bFf5sLavfOKRtqM/+1pWmcsTAnHrR/3izvlHXrN2fdhncfS1as3er5ps1N7+u+VRWlLUWT3xZQKt8toPzXQkp1RfZuV5WhB9alXCkFAAAAAAAA2AGUSwDep7LyitT07J8k6ZOkT/e9k+HX54j3+fubNjdtVTZZt35z1m7YtOV43YbNWft7x797rF2/OW+tXp/Fb/7++cat3vtrFx+VQf27bt8BAwAAAAAAAES5BGCnqawopbKiKu3bVn3g92pqas76jb8tmjQ2NadrhzbbISEAAAAAAADAtpRLAHZBpVJZ2tRUpk1NZdFRAAAAAAAAgN1cqegAAAAAAAAAAAC0XsolAAAAAAAAAAC0SLkEAAAAAAAAAIAWKZcAAAAAAAAAANAi5RIAAAAAAAAAAFqkXAIAAAAAAAAAQIsqig6wvTQ2NiZJGhoaCk4CsOv43Xfm775DAQAAAAAAAP6r3aZcsmzZsiTJeeedV3ASgF3PsmXLsu+++xYdAwAAAAAAAGiFypqbm5uLDrE9rF+/PrNmzUqXLl1SXl5edJw/SUNDQ84777zcc889qaurKzpO4czHu8zF1szH1j7IfDQ2NmbZsmU5+OCDU1NTs4MSAgAAAAAAALuy3WblkpqamgwZMqToGNtFXV1devbsWXSMVsN8vMtcbM18bO1PnQ8rlgAAAAAAAAB/SKnoAAAAAAAAAAAAtF7KJQAAAAAAAAAAtEi5BAAAAAAAAACAFpVfd9111xUdgndVV1dn2LBhqa6uLjpKq2A+3mUutmY+tmY+AAAAAAAAgB2lrLm5ubnoEAAAAAAAAAAAtE62xQEAAAAAAAAAoEXKJQAAAAAAAAAAtEi5pBVYuXJlLrroopx00kk59dRTc9lll2XFihVFx2oVbrnllvTv3z8vvfRS0VEKtWHDhlx77bU58cQTc+qpp+arX/1q0ZEK8/jjj+f000/PmDFjctppp2XatGlFR9qpxo8fn5EjR27zuZg3b17Gjh2bk046KWPHjs38+fOLCwkAAAAAAADsVpRLWoGysrJceOGFeeSRR/LQQw+lV69eufHGG4uOVbjZs2fnV7/6VXr06FF0lML93d/9Xaqrq7f8jVxxxRVFRypEc3NzvvCFL+SGG27IpEmTcsMNN+Tqq69OU1NT0dF2mlGjRuWee+7Z5nNx7bXX5txzz80jjzySc889N9dcc01BCQEAAAAAAIDdjXJJK7DPPvtk2LBhW54ffvjheeONNwpMVLyNGzfma1/7Wq677rqioxRuzZo1eeCBB3LFFVekrKwsSdK5c+eCUxWnVCpl9erVSZLVq1ena9euKZX2nK+yIUOGpL6+fqtzy5cvz5w5czJ69OgkyejRozNnzhwrIAEAAAAAAADbRUXRAdhaU1NTfvjDH2bkyJFFRynUt7/97Zx22mnp2bNn0VEKt3Dhwuyzzz655ZZb8swzz6Rt27a54oorMmTIkKKj7XRlZWW5+eabc+mll6ZNmzZZs2ZN7rrrrqJjFW7x4sXp1q1bysvLkyTl5eXp2rVrFi9enI4dOxacDgAAAAAAANjV7Tn/3X8X8fWvfz1t2rTJJz7xiaKjFGbGjBmZNWtWzj333KKjtAqNjY1ZuHBhDjzwwNx///256qqrMm7cuLzzzjtFR9vpNm/enDvvvDO33XZbHn/88dx+++258sors2bNmqKjAQAAAAAAAOy2lEtakfHjx2fBggW5+eab96htPv6r5557LnPnzs2oUaMycuTINDQ05IILLshTTz1VdLRC1NfXp6KiYsuWJ4cddlg6dOiQefPmFZxs53vhhReydOnSDB48OEkyePDg1NbWZu7cuQUnK1Z9fX2WLFmSxsbGJL8tJC1dunSb7XMAAAAAAAAA/hR7boOhlbnpppsya9as3Hrrramqqio6TqEuvvjiPPXUU3nsscfy2GOPpa6uLv/4j/+YY445puhohejYsWOGDRuW6dOnJ0nmzZuX5cuXZ9999y042c5XV1eXhoaGvPrqq0mSuXPnZvny5endu3fByYrVqVOnDBw4MJMnT06STJ48OQMHDrQlDgAAAAAAALBdlDU3NzcXHWJP9/LLL2f06NHZb7/9UlNTkyTp2bNnbr311oKTtQ4jR47MHXfckX79+hUdpTALFy7Ml770pbz11lupqKjIlVdemWOPPbboWIV48MEH853vfCdlZWVJkssvvzwnnHBCwal2nm984xuZNm1a3nzzzXTo0CH77LNPpkyZkrlz5+aLX/xiVq1alfbt22f8+PHZf//9i44LAAAAAAAA7AaUSwAAAAAAAAAAaJFtcQAAAAAAAAAAaJFyCQAAAAAAAAAALVIuAQAAAAAAAACgRcolAAAAAAAAAAC0SLkEAAAAAAAAAIAWKZdAK/P666+nf//+2bx5c9FRAAAAAAAAAEC5BAAAAAAAAACAlimXAAAAAAAAAADQIuUSeB+WLFmScePGZfjw4Rk5cmTuvvvuJMmECRNy+eWX58orr8ygQYNyxhln5MUXX9zye3Pnzs0nP/nJDBkyJKecckoeffTRLa+tX78+3/rWt3L88cdn8ODBOeecc7J+/fotrz/00EM57rjjMmzYsNx+++07b7AAAAAAAAAA8HuUS+A9NDU15bOf/Wz69++fJ554IhMnTszEiRPz5JNPJkkeffTRnHzyyXn22WczevToXHrppdm0aVM2bdqUSy65JCNGjMjPfvazfOUrX8lVV12VV199NUkyfvz4zJ49O/fee2+effbZfP7zn0+p9O5H8he/+EUefvjhTJw4Mbfeemvmzp1byPgBAAAAAAAA2LMpl8B7eP7557NixYpcdtllqaqqSq9evfKxj30sU6dOTZIcdNBBOfnkk1NZWZnzzz8/GzduzMyZMzNz5sysXbs2F198caqqqnLUUUfl+OOPz5QpU9LU1JQf/ehH+fKXv5xu3bqlvLw8RxxxRKqqqrbc97LLLktNTU0GDBiQAQMGbLUiCgAAAAAAAADsLBVFB4DWbtGiRVm6dGmGDBmy5VxjY2OGDBmS7t27p66ubsv5UqmUbt26ZenSpUmSurq6rVYj6d69e5YsWZKVK1dmw4YN6dWrV4v37dy585bj2trarF27dnsOCwAAAAAAAADeF+USeA/19fXp2bNnpk2bts1rEyZMSENDw5bnTU1NWbJkSbp27ZokaWhoSFNT05aCyeLFi7PffvulQ4cOqa6uzsKFCzNgwICdMxAAAAAAAAAA+BPYFgfew6GHHpq2bdvmrrvuyvr169PY2JiXXnopv/71r5Mks2fPzrRp07J58+ZMnDgxVVVVOeyww3LooYempqYm3/3ud7Np06Y888wzeeyxx/LRj340pVIpZ511Vr75zW9myZIlaWxszIwZM7Jx48aCRwsAAAAAAAAAW1MugfdQXl6eO+64Iy+++GJGjRqV4cOH5ytf+UreeeedJMmoUaMyderUHHnkkZk0aVImTJiQysrKVFVV5Y477sgTTzyR4cOH5/rrr88NN9yQvn37Jkmuvvrq9OvXL2effXaGDh2aG2+8MU1NTUUOFQAAAAAAAAC2Udbc3NxcdAjYVU2YMCELFizIjTfeWHQUAAAAAAAAANghrFwCAAAAAAAAAECLlEsAAAAAAAAAAGiRbXEAAAAAAAAAAGiRlUsAAAAAAAAAAGiRcgkAAAAAAAAAAC1SLgEAAAAAAAAAoEXKJQAAAAAAAAAAtEi5BAAAAAAAAACAFimXAAAAAAAAAADQov8HzZ50wcxkypYAAAAASUVORK5CYII=\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {}
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "PvGetAcBH0Kw"
+      },
+      "source": [
+        "**Gradient Boosting Decision Tree Ensemble (GBDT)**"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "wzwdbWHGI9gQ",
+        "outputId": "d11d5fd5-6577-478d-ec40-cb9383ba41eb"
+      },
+      "source": [
+        "import os\n",
+        "import time\n",
+        "from bayes_opt import BayesianOptimization\n",
+        "#from sklearn.model_selection import cross_val_score\n",
+        "from sklearn.model_selection import train_test_split\n",
+        "from sklearn.metrics import mean_squared_error\n",
+        "from lightgbm import LGBMRegressor\n",
+        "import numpy as np\n",
+        "#import seaborn as sns\n",
+        "#from sklearn.preprocessing import MinMaxScaler\n",
+        "from sklearn.metrics import explained_variance_score\n",
+        "#from sklearn.preprocessing import StandardScaler\n",
+        "import matplotlib.pyplot as plt\n",
+        "import datetime\n",
+        "import pandas as pd\n",
+        "\n",
+        "t = time.localtime()\n",
+        "model_name = 'Invar_inference_GBDT'\n",
+        "file_name = '{}.xlsx'.format(model_name)\n",
+        "data = pd.read_csv('data_base.csv')\n",
+        "train_features, test_features, train_labels, test_labels = normalizing_data(data,seed=42)\n",
+        "train_features, test_features = train_features.cpu().data.numpy(),test_features.cpu().data.numpy()\n",
+        "train_labels, test_labels = train_labels.cpu().data.numpy(), test_labels.cpu().data.numpy()\n",
+        "train_labels, test_labels = train_labels.reshape(-1), test_labels.reshape(-1) \n",
+        "def train_model(num_leaves,\n",
+        "                min_child_samples,\n",
+        "            learning_rate,\n",
+        "            n_estimators, \n",
+        "            max_bin,\n",
+        "            colsample_bytree, \n",
+        "            subsample, \n",
+        "            max_depth, \n",
+        "            reg_alpha,\n",
+        "            reg_lambda,\n",
+        "            min_split_gain,\n",
+        "            min_child_weight\n",
+        "            ):\n",
+        "    params = {\n",
+        "        \"num_leaves\": int(round(num_leaves)),\n",
+        "        'min_child_samples':int(round(min_child_samples)),\n",
+        "        'learning_rate': learning_rate,\n",
+        "        'n_estimators': int(round(n_estimators)),\n",
+        "        'max_bin': int(round(max_bin)),\n",
+        "        'colsample_bytree': max(min(colsample_bytree, 1), 0),\n",
+        "        'subsample': max(min(subsample, 1), 0),\n",
+        "        'max_depth': int(round(max_depth)),\n",
+        "        'reg_alpha':  max(reg_alpha, 0),\n",
+        "        'reg_lambda': max(reg_lambda, 0),\n",
+        "        'min_split_gain': min_split_gain,\n",
+        "        'min_child_weight': min_child_weight,\n",
+        "        'verbose': -1\n",
+        "                  }\n",
+        "    model = LGBMRegressor(**params)\n",
+        "    model.fit(train_features, train_labels)\n",
+        "    y_pred = model.predict(test_features)\n",
+        "    error = -np.mean(np.abs((test_labels - y_pred) / test_labels))       # print(error)     \n",
+        "    return error\n",
+        "bounds = {'num_leaves': (5, 60),#50\n",
+        "          'min_child_samples':(1, 50),\n",
+        "          'learning_rate': (0.001, 1),\n",
+        "          'n_estimators': (5, 200),#100\n",
+        "            'max_bin': (5, 100),#10\n",
+        "          'colsample_bytree': (0.5, 1),\n",
+        "          'subsample': (0.1, 2),\n",
+        "          'max_depth': (1, 60),#10\n",
+        "          'reg_alpha': (0.01, 1), #5\n",
+        "          'reg_lambda': (0.01, 1),#5\n",
+        "          'min_split_gain': (0.001, 0.1),\n",
+        "          'min_child_weight': (0.0001, 30)}\n",
+        "optimizer = BayesianOptimization(\n",
+        "    f=train_model,\n",
+        "    pbounds=bounds,\n",
+        "    random_state=1,\n",
+        ")\n",
+        "optimizer.maximize(init_points = 10, n_iter=1)\n",
+        "table = pd.DataFrame(columns=['target', 'colsample_bytree', 'learning_rate', 'max_bin',\n",
+        "                      'max_depth','min_child_samples','min_child_weight','min_split_gain',\n",
+        "                      'n_estimators','num_leaves','reg_alpha','reg_lambda','subsample'])\n",
+        "for res in optimizer.res:\n",
+        "    table=table.append(pd.DataFrame({'target':[res['target']],'colsample_bytree':[res['params']['colsample_bytree']],\n",
+        "                                     'colsample_bytree':[res['params']['colsample_bytree']],\n",
+        "                                     'learning_rate':[res['params']['learning_rate']],\n",
+        "                                     'max_bin':[res['params']['max_bin']],\n",
+        "                                     'max_depth':[res['params']['max_depth']],\n",
+        "                                     'min_child_samples':[res['params']['min_child_samples']],\n",
+        "                                     'min_child_weight':[res['params']['min_child_weight']],\n",
+        "                                     'min_split_gain':[res['params']['min_split_gain']],\n",
+        "                                     'n_estimators':[res['params']['n_estimators']],\n",
+        "                                     'num_leaves':[res['params']['num_leaves']],\n",
+        "                                     'reg_alpha':[res['params']['reg_alpha']],\n",
+        "                                     'reg_lambda':[res['params']['reg_lambda']],\n",
+        "                                     'subsample':[res['params']['subsample']]}),\n",
+        "                                     ignore_index=True)\n",
+        "table=table.append(pd.DataFrame({'target':[optimizer.max['target']],'colsample_bytree':[optimizer.max['params']['colsample_bytree']],\n",
+        "                                 'colsample_bytree':[optimizer.max['params']['colsample_bytree']],\n",
+        "                                 'learning_rate':[optimizer.max['params']['learning_rate']],\n",
+        "                                 'max_bin':[optimizer.max['params']['max_bin']],\n",
+        "                                 'max_depth':[optimizer.max['params']['max_depth']],\n",
+        "                                 'min_child_samples':[optimizer.max['params']['min_child_samples']],\n",
+        "                                 'min_child_weight':[optimizer.max['params']['min_child_weight']],\n",
+        "                                 'min_split_gain':[optimizer.max['params']['min_split_gain']],\n",
+        "                                 'n_estimators':[optimizer.max['params']['n_estimators']],\n",
+        "                                 'num_leaves':[optimizer.max['params']['num_leaves']],\n",
+        "                                 'reg_alpha':[optimizer.max['params']['reg_alpha']],\n",
+        "                                 'reg_lambda':[optimizer.max['params']['reg_lambda']],\n",
+        "                                 'subsample':[optimizer.max['params']['subsample']]}),\n",
+        "                                 ignore_index=True)\n",
+        "table.to_excel(file_name)\n",
+        "endtime = datetime.datetime.now()\n",
+        "print ('running time {}'.format(endtime - starttime))"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "|   iter    |  target   | colsam... | learni... |  max_bin  | max_depth | min_ch... | min_ch... | min_sp... | n_esti... | num_le... | reg_alpha | reg_la... | subsample |\n",
+            "-------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n",
+            "| \u001b[0m 1       \u001b[0m | \u001b[0m-0.294   \u001b[0m | \u001b[0m 0.7085  \u001b[0m | \u001b[0m 0.7206  \u001b[0m | \u001b[0m 5.011   \u001b[0m | \u001b[0m 18.84   \u001b[0m | \u001b[0m 8.191   \u001b[0m | \u001b[0m 2.77    \u001b[0m | \u001b[0m 0.01944 \u001b[0m | \u001b[0m 72.38   \u001b[0m | \u001b[0m 26.82   \u001b[0m | \u001b[0m 0.5434  \u001b[0m | \u001b[0m 0.425   \u001b[0m | \u001b[0m 1.402   \u001b[0m |\n",
+            "| \u001b[0m 2       \u001b[0m | \u001b[0m-0.5621  \u001b[0m | \u001b[0m 0.6022  \u001b[0m | \u001b[0m 0.8782  \u001b[0m | \u001b[0m 7.602   \u001b[0m | \u001b[0m 40.56   \u001b[0m | \u001b[0m 21.45   \u001b[0m | \u001b[0m 16.76   \u001b[0m | \u001b[0m 0.0149  \u001b[0m | \u001b[0m 43.63   \u001b[0m | \u001b[0m 49.04   \u001b[0m | \u001b[0m 0.9686  \u001b[0m | \u001b[0m 0.3203  \u001b[0m | \u001b[0m 1.415   \u001b[0m |\n",
+            "| \u001b[95m 3       \u001b[0m | \u001b[95m-0.2718  \u001b[0m | \u001b[95m 0.9382  \u001b[0m | \u001b[95m 0.8947  \u001b[0m | \u001b[95m 13.08   \u001b[0m | \u001b[95m 3.304   \u001b[0m | \u001b[95m 9.322   \u001b[0m | \u001b[95m 26.34   \u001b[0m | \u001b[95m 0.01074 \u001b[0m | \u001b[95m 87.12   \u001b[0m | \u001b[95m 57.68   \u001b[0m | \u001b[95m 0.5378  \u001b[0m | \u001b[95m 0.695   \u001b[0m | \u001b[95m 0.6995  \u001b[0m |\n",
+            "| \u001b[0m 4       \u001b[0m | \u001b[0m-0.3307  \u001b[0m | \u001b[0m 0.8433  \u001b[0m | \u001b[0m 0.8348  \u001b[0m | \u001b[0m 6.737   \u001b[0m | \u001b[0m 45.26   \u001b[0m | \u001b[0m 49.45   \u001b[0m | \u001b[0m 22.44   \u001b[0m | \u001b[0m 0.02876 \u001b[0m | \u001b[0m 158.9   \u001b[0m | \u001b[0m 10.68   \u001b[0m | \u001b[0m 0.4534  \u001b[0m | \u001b[0m 0.9095  \u001b[0m | \u001b[0m 0.6579  \u001b[0m |\n",
+            "| \u001b[0m 5       \u001b[0m | \u001b[0m-0.5333  \u001b[0m | \u001b[0m 0.6439  \u001b[0m | \u001b[0m 0.1309  \u001b[0m | \u001b[0m 6.84    \u001b[0m | \u001b[0m 41.05   \u001b[0m | \u001b[0m 11.37   \u001b[0m | \u001b[0m 7.966   \u001b[0m | \u001b[0m 0.04967 \u001b[0m | \u001b[0m 15.41   \u001b[0m | \u001b[0m 36.58   \u001b[0m | \u001b[0m 0.1553  \u001b[0m | \u001b[0m 0.5934  \u001b[0m | \u001b[0m 1.43    \u001b[0m |\n",
+            "| \u001b[95m 6       \u001b[0m | \u001b[95m-0.1948  \u001b[0m | \u001b[95m 0.5512  \u001b[0m | \u001b[95m 0.4146  \u001b[0m | \u001b[95m 70.97   \u001b[0m | \u001b[95m 25.44   \u001b[0m | \u001b[95m 3.448   \u001b[0m | \u001b[95m 16.08   \u001b[0m | \u001b[95m 0.06672 \u001b[0m | \u001b[95m 105.4   \u001b[0m | \u001b[95m 56.95   \u001b[0m | \u001b[95m 0.5907  \u001b[0m | \u001b[95m 0.9044  \u001b[0m | \u001b[95m 0.3612  \u001b[0m |\n",
+            "| \u001b[0m 7       \u001b[0m | \u001b[0m-0.278   \u001b[0m | \u001b[0m 0.5696  \u001b[0m | \u001b[0m 0.8076  \u001b[0m | \u001b[0m 42.78   \u001b[0m | \u001b[0m 10.76   \u001b[0m | \u001b[0m 46.45   \u001b[0m | \u001b[0m 10.43   \u001b[0m | \u001b[0m 0.07533 \u001b[0m | \u001b[0m 146.6   \u001b[0m | \u001b[0m 53.58   \u001b[0m | \u001b[0m 0.6274  \u001b[0m | \u001b[0m 0.7534  \u001b[0m | \u001b[0m 0.7629  \u001b[0m |\n",
+            "| \u001b[0m 8       \u001b[0m | \u001b[0m-0.2663  \u001b[0m | \u001b[0m 0.635   \u001b[0m | \u001b[0m 0.896   \u001b[0m | \u001b[0m 45.67   \u001b[0m | \u001b[0m 57.93   \u001b[0m | \u001b[0m 33.51   \u001b[0m | \u001b[0m 18.65   \u001b[0m | \u001b[0m 0.01236 \u001b[0m | \u001b[0m 190.2   \u001b[0m | \u001b[0m 29.75   \u001b[0m | \u001b[0m 0.5826  \u001b[0m | \u001b[0m 0.4141  \u001b[0m | \u001b[0m 0.5504  \u001b[0m |\n",
+            "| \u001b[0m 9       \u001b[0m | \u001b[0m-0.292   \u001b[0m | \u001b[0m 0.9517  \u001b[0m | \u001b[0m 0.5741  \u001b[0m | \u001b[0m 5.273   \u001b[0m | \u001b[0m 37.41   \u001b[0m | \u001b[0m 17.01   \u001b[0m | \u001b[0m 15.81   \u001b[0m | \u001b[0m 0.08871 \u001b[0m | \u001b[0m 74.67   \u001b[0m | \u001b[0m 54.97   \u001b[0m | \u001b[0m 0.6271  \u001b[0m | \u001b[0m 0.02566 \u001b[0m | \u001b[0m 1.866   \u001b[0m |\n",
+            "| \u001b[0m 10      \u001b[0m | \u001b[0m-0.2683  \u001b[0m | \u001b[0m 0.8454  \u001b[0m | \u001b[0m 0.9973  \u001b[0m | \u001b[0m 21.37   \u001b[0m | \u001b[0m 9.091   \u001b[0m | \u001b[0m 46.7    \u001b[0m | \u001b[0m 20.9    \u001b[0m | \u001b[0m 0.007534\u001b[0m | \u001b[0m 152.3   \u001b[0m | \u001b[0m 46.46   \u001b[0m | \u001b[0m 0.9238  \u001b[0m | \u001b[0m 0.7144  \u001b[0m | \u001b[0m 0.3361  \u001b[0m |\n",
+            "| \u001b[0m 11      \u001b[0m | \u001b[0m-1.377   \u001b[0m | \u001b[0m 1.0     \u001b[0m | \u001b[0m 0.001   \u001b[0m | \u001b[0m 5.0     \u001b[0m | \u001b[0m 1.0     \u001b[0m | \u001b[0m 1.0     \u001b[0m | \u001b[0m 0.0001  \u001b[0m | \u001b[0m 0.1     \u001b[0m | \u001b[0m 200.0   \u001b[0m | \u001b[0m 60.0    \u001b[0m | \u001b[0m 0.01    \u001b[0m | \u001b[0m 0.01    \u001b[0m | \u001b[0m 2.0     \u001b[0m |\n",
+            "=========================================================================================================================================================================\n",
+            "running time 0:03:25.836239\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "o3mxAtKuIHO0"
+      },
+      "source": [
+        "**Introducing the atomic properties**\n",
+        "\n",
+        "All were found in the standard text book.\n",
+        "e.g., valence electron concentration(VEC)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "3dschQ9eKwNY"
+      },
+      "source": [
+        "def atomic_properties(new_comps):\n",
+        "  df_test = new_comps.copy()\n",
+        "  df_test['VEC'] = 8*df_test['Fe'] + 10*df_test['Ni'] + 9*df_test['Co'] + 6*df_test['Cr'] + 5*df_test['V'] + 11*df_test['Cu']\n",
+        "  df_test['AR1'] = 140*df_test['Fe'] + 135*df_test['Ni'] + 135*df_test['Co'] + 140*df_test['Cr'] + 135*df_test['V'] + 135*df_test['Cu']\n",
+        "  df_test['AR2'] = 124*df_test['Fe'] + 125*df_test['Ni'] + 125*df_test['Co'] + 125*df_test['Cr'] + 132*df_test['V'] + 128*df_test['Cu']\n",
+        "  df_test['PE'] = 1.83*df_test['Fe'] + 1.91*df_test['Ni'] + 1.88*df_test['Co'] + 1.66*df_test['Cr'] + 1.63*df_test['V'] + 1.9*df_test['Cu']\n",
+        "  df_test['Density'] = 7874*df_test['Fe'] + 8908*df_test['Ni'] + 8900*df_test['Co'] + 7140*df_test['Cr'] + 6110*df_test['V'] + 8920*df_test['Cu']\n",
+        "  df_test['TC'] = 80*df_test['Fe'] + 91*df_test['Ni'] + 100*df_test['Co'] + 94*df_test['Cr'] + 30.7*df_test['V'] + 400*df_test['Cu']\n",
+        "  df_test['MP'] = 1181*df_test['Fe'] + 1728*df_test['Ni'] + 1768*df_test['Co'] + 2180*df_test['Cr'] + 2183*df_test['V'] + 1357.77*df_test['Cu']\n",
+        "  df_test['FI'] = 762.47*df_test['Fe'] + 737.14*df_test['Ni'] + 760.4*df_test['Co'] + 652.87*df_test['Cr'] + 650.91*df_test['V'] + 745.78*df_test['Cu']\n",
+        "  df_test['SI'] = 1562.98*df_test['Fe'] + 1753.03*df_test['Ni'] + 1648.39*df_test['Co'] + 1590.69*df_test['Cr'] + 1412*df_test['V'] + 1957.92*df_test['Cu']\n",
+        "  df_test['TI'] = 2957.4*df_test['Fe'] + 3395*df_test['Ni'] + 3232.3*df_test['Co'] + 2987.1*df_test['Cr'] + 2828.09*df_test['V'] + 3554.6*df_test['Cu']\n",
+        "  df_test['M'] = 2.22*df_test['Fe'] + 0.6*df_test['Ni'] + 1.72*df_test['Co'] + -0.6*df_test['Cr'] + 0.0*df_test['V'] + 0.0*df_test['Cu']\n",
+        "  return df_test"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "bJAo_7g4L5BW"
+      },
+      "source": [
+        "WAE_comps=pd.read_csv('comps_WAE.csv')"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 203
+        },
+        "id": "WP4-WdrdMCr1",
+        "outputId": "2db4a3a9-d1d0-43e1-a3fb-f4ffe432c00f"
+      },
+      "source": [
+        "WAE_comps.head()"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>Fe</th>\n",
+              "      <th>Ni</th>\n",
+              "      <th>Co</th>\n",
+              "      <th>Cr</th>\n",
+              "      <th>V</th>\n",
+              "      <th>Cu</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>0.014077</td>\n",
+              "      <td>0.231521</td>\n",
+              "      <td>0.754396</td>\n",
+              "      <td>1.664660e-06</td>\n",
+              "      <td>0.000004</td>\n",
+              "      <td>1.433623e-07</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>0.141357</td>\n",
+              "      <td>0.333938</td>\n",
+              "      <td>0.524631</td>\n",
+              "      <td>7.136076e-07</td>\n",
+              "      <td>0.000072</td>\n",
+              "      <td>3.349655e-07</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>0.271059</td>\n",
+              "      <td>0.108174</td>\n",
+              "      <td>0.540923</td>\n",
+              "      <td>8.701057e-07</td>\n",
+              "      <td>0.079841</td>\n",
+              "      <td>1.835431e-06</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>0.274571</td>\n",
+              "      <td>0.086473</td>\n",
+              "      <td>0.587006</td>\n",
+              "      <td>1.669761e-06</td>\n",
+              "      <td>0.051947</td>\n",
+              "      <td>1.217349e-06</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>0.461331</td>\n",
+              "      <td>0.259892</td>\n",
+              "      <td>0.183169</td>\n",
+              "      <td>2.289097e-06</td>\n",
+              "      <td>0.095604</td>\n",
+              "      <td>2.271647e-06</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "         Fe        Ni        Co            Cr         V            Cu\n",
+              "0  0.014077  0.231521  0.754396  1.664660e-06  0.000004  1.433623e-07\n",
+              "1  0.141357  0.333938  0.524631  7.136076e-07  0.000072  3.349655e-07\n",
+              "2  0.271059  0.108174  0.540923  8.701057e-07  0.079841  1.835431e-06\n",
+              "3  0.274571  0.086473  0.587006  1.669761e-06  0.051947  1.217349e-06\n",
+              "4  0.461331  0.259892  0.183169  2.289097e-06  0.095604  2.271647e-06"
+            ]
+          },
+          "metadata": {},
+          "execution_count": 52
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "picNF-1gLuPN"
+      },
+      "source": [
+        "WAE_comps=atomic_properties(WAE_comps)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 307
+        },
+        "id": "SzEg5zz0MdPu",
+        "outputId": "3a1e4792-15fc-48c0-f70d-73ae126d6f9f"
+      },
+      "source": [
+        "WAE_comps.head()"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>Fe</th>\n",
+              "      <th>Ni</th>\n",
+              "      <th>Co</th>\n",
+              "      <th>Cr</th>\n",
+              "      <th>V</th>\n",
+              "      <th>Cu</th>\n",
+              "      <th>VEC</th>\n",
+              "      <th>AR1</th>\n",
+              "      <th>AR2</th>\n",
+              "      <th>PE</th>\n",
+              "      <th>Density</th>\n",
+              "      <th>TC</th>\n",
+              "      <th>MP</th>\n",
+              "      <th>FI</th>\n",
+              "      <th>SI</th>\n",
+              "      <th>TI</th>\n",
+              "      <th>M</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>0.014077</td>\n",
+              "      <td>0.231521</td>\n",
+              "      <td>0.754396</td>\n",
+              "      <td>1.664660e-06</td>\n",
+              "      <td>0.000004</td>\n",
+              "      <td>1.433623e-07</td>\n",
+              "      <td>9.217423</td>\n",
+              "      <td>135.070392</td>\n",
+              "      <td>124.985951</td>\n",
+              "      <td>1.886240</td>\n",
+              "      <td>8887.395013</td>\n",
+              "      <td>97.634525</td>\n",
+              "      <td>1750.478257</td>\n",
+              "      <td>755.043333</td>\n",
+              "      <td>1671.413043</td>\n",
+              "      <td>3266.096719</td>\n",
+              "      <td>1.467724</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>0.141357</td>\n",
+              "      <td>0.333938</td>\n",
+              "      <td>0.524631</td>\n",
+              "      <td>7.136076e-07</td>\n",
+              "      <td>0.000072</td>\n",
+              "      <td>3.349655e-07</td>\n",
+              "      <td>9.192292</td>\n",
+              "      <td>135.706802</td>\n",
+              "      <td>124.859160</td>\n",
+              "      <td>1.882932</td>\n",
+              "      <td>8757.437080</td>\n",
+              "      <td>94.162511</td>\n",
+              "      <td>1671.695978</td>\n",
+              "      <td>752.917285</td>\n",
+              "      <td>1671.243138</td>\n",
+              "      <td>3247.743705</td>\n",
+              "      <td>1.416541</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>0.271059</td>\n",
+              "      <td>0.108174</td>\n",
+              "      <td>0.540923</td>\n",
+              "      <td>8.701057e-07</td>\n",
+              "      <td>0.079841</td>\n",
+              "      <td>1.835431e-06</td>\n",
+              "      <td>8.517753</td>\n",
+              "      <td>136.355311</td>\n",
+              "      <td>125.287840</td>\n",
+              "      <td>1.849732</td>\n",
+              "      <td>8400.001835</td>\n",
+              "      <td>88.072832</td>\n",
+              "      <td>1637.694757</td>\n",
+              "      <td>749.703133</td>\n",
+              "      <td>1617.685234</td>\n",
+              "      <td>3143.113863</td>\n",
+              "      <td>1.597044</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>0.274571</td>\n",
+              "      <td>0.086473</td>\n",
+              "      <td>0.587006</td>\n",
+              "      <td>1.669761e-06</td>\n",
+              "      <td>0.051947</td>\n",
+              "      <td>1.217349e-06</td>\n",
+              "      <td>8.604112</td>\n",
+              "      <td>136.372854</td>\n",
+              "      <td>125.089050</td>\n",
+              "      <td>1.855878</td>\n",
+              "      <td>8474.046874</td>\n",
+              "      <td>90.130761</td>\n",
+              "      <td>1624.925709</td>\n",
+              "      <td>753.269100</td>\n",
+              "      <td>1621.707889</td>\n",
+              "      <td>3149.891904</td>\n",
+              "      <td>1.671081</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>0.461331</td>\n",
+              "      <td>0.259892</td>\n",
+              "      <td>0.183169</td>\n",
+              "      <td>2.289097e-06</td>\n",
+              "      <td>0.095604</td>\n",
+              "      <td>2.271647e-06</td>\n",
+              "      <td>8.416143</td>\n",
+              "      <td>137.306652</td>\n",
+              "      <td>125.207892</td>\n",
+              "      <td>1.840829</td>\n",
+              "      <td>8162.014461</td>\n",
+              "      <td>81.809674</td>\n",
+              "      <td>1526.478738</td>\n",
+              "      <td>744.841857</td>\n",
+              "      <td>1613.583472</td>\n",
+              "      <td>3109.120486</td>\n",
+              "      <td>1.495138</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "         Fe        Ni        Co  ...           SI           TI         M\n",
+              "0  0.014077  0.231521  0.754396  ...  1671.413043  3266.096719  1.467724\n",
+              "1  0.141357  0.333938  0.524631  ...  1671.243138  3247.743705  1.416541\n",
+              "2  0.271059  0.108174  0.540923  ...  1617.685234  3143.113863  1.597044\n",
+              "3  0.274571  0.086473  0.587006  ...  1621.707889  3149.891904  1.671081\n",
+              "4  0.461331  0.259892  0.183169  ...  1613.583472  3109.120486  1.495138\n",
+              "\n",
+              "[5 rows x 17 columns]"
+            ]
+          },
+          "metadata": {},
+          "execution_count": 54
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "TRInGf1JImc-"
+      },
+      "source": [
+        "We want the atomic properties to be normalized (so that it can be understood by neural networks)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "Sg4m9Q25MZqk"
+      },
+      "source": [
+        "composition= WAE_comps[['Fe','Ni','Co','Cr','V','Cu']]\n",
+        "min_max_scaler = preprocessing.MinMaxScaler()\n",
+        "normalized_atomic_properties = min_max_scaler.fit_transform(WAE_comps[['VEC','AR1','AR2','PE','Density',\n",
+        "                                              'TC','MP','FI','SI','TI','M']])\n",
+        "WAE_x = pd.concat([composition,pd.DataFrame(normalized_atomic_properties)],axis=1)"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 307
+        },
+        "id": "29QZZChOM1Y5",
+        "outputId": "21377392-bb83-4654-e12c-c47ddbc77703"
+      },
+      "source": [
+        "WAE_x.head()"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>Fe</th>\n",
+              "      <th>Ni</th>\n",
+              "      <th>Co</th>\n",
+              "      <th>Cr</th>\n",
+              "      <th>V</th>\n",
+              "      <th>Cu</th>\n",
+              "      <th>0</th>\n",
+              "      <th>1</th>\n",
+              "      <th>2</th>\n",
+              "      <th>3</th>\n",
+              "      <th>4</th>\n",
+              "      <th>5</th>\n",
+              "      <th>6</th>\n",
+              "      <th>7</th>\n",
+              "      <th>8</th>\n",
+              "      <th>9</th>\n",
+              "      <th>10</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>0.014077</td>\n",
+              "      <td>0.231521</td>\n",
+              "      <td>0.754396</td>\n",
+              "      <td>1.664660e-06</td>\n",
+              "      <td>0.000004</td>\n",
+              "      <td>1.433623e-07</td>\n",
+              "      <td>0.608974</td>\n",
+              "      <td>0.013980</td>\n",
+              "      <td>0.655022</td>\n",
+              "      <td>0.718451</td>\n",
+              "      <td>0.980616</td>\n",
+              "      <td>0.581757</td>\n",
+              "      <td>0.850225</td>\n",
+              "      <td>0.707755</td>\n",
+              "      <td>0.570637</td>\n",
+              "      <td>0.705669</td>\n",
+              "      <td>0.535856</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>0.141357</td>\n",
+              "      <td>0.333938</td>\n",
+              "      <td>0.524631</td>\n",
+              "      <td>7.136076e-07</td>\n",
+              "      <td>0.000072</td>\n",
+              "      <td>3.349655e-07</td>\n",
+              "      <td>0.596393</td>\n",
+              "      <td>0.141387</td>\n",
+              "      <td>0.570666</td>\n",
+              "      <td>0.679149</td>\n",
+              "      <td>0.854777</td>\n",
+              "      <td>0.485289</td>\n",
+              "      <td>0.732468</td>\n",
+              "      <td>0.623713</td>\n",
+              "      <td>0.569742</td>\n",
+              "      <td>0.663670</td>\n",
+              "      <td>0.504209</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>0.271059</td>\n",
+              "      <td>0.108174</td>\n",
+              "      <td>0.540923</td>\n",
+              "      <td>8.701057e-07</td>\n",
+              "      <td>0.079841</td>\n",
+              "      <td>1.835431e-06</td>\n",
+              "      <td>0.258708</td>\n",
+              "      <td>0.271216</td>\n",
+              "      <td>0.855873</td>\n",
+              "      <td>0.284718</td>\n",
+              "      <td>0.508671</td>\n",
+              "      <td>0.316090</td>\n",
+              "      <td>0.681646</td>\n",
+              "      <td>0.496659</td>\n",
+              "      <td>0.287527</td>\n",
+              "      <td>0.424237</td>\n",
+              "      <td>0.615817</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>0.274571</td>\n",
+              "      <td>0.086473</td>\n",
+              "      <td>0.587006</td>\n",
+              "      <td>1.669761e-06</td>\n",
+              "      <td>0.051947</td>\n",
+              "      <td>1.217349e-06</td>\n",
+              "      <td>0.301941</td>\n",
+              "      <td>0.274728</td>\n",
+              "      <td>0.723615</td>\n",
+              "      <td>0.357740</td>\n",
+              "      <td>0.580369</td>\n",
+              "      <td>0.373269</td>\n",
+              "      <td>0.662559</td>\n",
+              "      <td>0.637620</td>\n",
+              "      <td>0.308724</td>\n",
+              "      <td>0.439748</td>\n",
+              "      <td>0.661596</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>0.461331</td>\n",
+              "      <td>0.259892</td>\n",
+              "      <td>0.183169</td>\n",
+              "      <td>2.289097e-06</td>\n",
+              "      <td>0.095604</td>\n",
+              "      <td>2.271647e-06</td>\n",
+              "      <td>0.207841</td>\n",
+              "      <td>0.461670</td>\n",
+              "      <td>0.802682</td>\n",
+              "      <td>0.178942</td>\n",
+              "      <td>0.278227</td>\n",
+              "      <td>0.142071</td>\n",
+              "      <td>0.515409</td>\n",
+              "      <td>0.304494</td>\n",
+              "      <td>0.265913</td>\n",
+              "      <td>0.346447</td>\n",
+              "      <td>0.552807</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "         Fe        Ni        Co  ...         8         9        10\n",
+              "0  0.014077  0.231521  0.754396  ...  0.570637  0.705669  0.535856\n",
+              "1  0.141357  0.333938  0.524631  ...  0.569742  0.663670  0.504209\n",
+              "2  0.271059  0.108174  0.540923  ...  0.287527  0.424237  0.615817\n",
+              "3  0.274571  0.086473  0.587006  ...  0.308724  0.439748  0.661596\n",
+              "4  0.461331  0.259892  0.183169  ...  0.265913  0.346447  0.552807\n",
+              "\n",
+              "[5 rows x 17 columns]"
+            ]
+          },
+          "metadata": {},
+          "execution_count": 56
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "sYiCzAquIyVv"
+      },
+      "source": [
+        "**The final ensemble model**\n",
+        "\n",
+        "the final ensembles combine 4 GBDT and 4 NN to predict TEC of the WAE-generated compositions."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "hvhd70hpK4_J",
+        "outputId": "97c91cfc-f711-4d21-de35-d77eaf6b85fc"
+      },
+      "source": [
+        "def Tree(n,j, WAE_x):\n",
+        "    target = pd.read_excel('Invar_inference_GBDT.xlsx')\n",
+        "    colsample_bytree = target.at[n,'colsample_bytree']\n",
+        "    learning_rate = target.at[n,'learning_rate']\n",
+        "    max_bin = target.at[n,'max_bin']\n",
+        "    max_depth = target.at[n,'max_depth']\n",
+        "    max_bin = target.at[n,'max_bin']\n",
+        "    min_child_samples = target.at[n,'min_child_samples']\n",
+        "    min_child_weight = target.at[n,'min_child_weight']\n",
+        "    min_split_gain= target.at[n,'min_split_gain']\n",
+        "    n_estimators = target.at[n,'n_estimators']\n",
+        "    num_leaves = target.at[n,'num_leaves']\n",
+        "    reg_alpha = target.at[n,'reg_alpha']\n",
+        "    reg_lambda = target.at[n,'reg_lambda']\n",
+        "    subsample = target.at[n,'subsample']\n",
+        "    params = {\n",
+        "        \"num_leaves\": int(round(num_leaves)),\n",
+        "        'min_child_samples':int(round(min_child_samples)),\n",
+        "        'learning_rate': learning_rate,\n",
+        "        'n_estimators': int(round(n_estimators)),\n",
+        "        'max_bin': int(round(max_bin)),\n",
+        "        'colsample_bytree': max(min(colsample_bytree, 1), 0),\n",
+        "        'subsample': max(min(subsample, 1), 0),\n",
+        "        'max_depth': int(round(max_depth)),\n",
+        "        'reg_lambda':  max(reg_lambda, 0),\n",
+        "        'reg_alpha': max(reg_alpha, 0),\n",
+        "        'min_split_gain': min_split_gain,\n",
+        "        'min_child_weight': min_child_weight,\n",
+        "        'objective': 'regression',\n",
+        "        'verbose': -1\n",
+        "                 }\n",
+        "    data=pd.read_csv('data_base.csv')             \n",
+        "    train_features, test_features, train_labels, test_labels = normalizing_data(data,seed=j)\n",
+        "    train_features, test_features = train_features.cpu().data.numpy(),test_features.cpu().data.numpy()\n",
+        "    train_labels, test_labels = train_labels.cpu().data.numpy(), test_labels.cpu().data.numpy()\n",
+        "    train_labels, test_labels = train_labels.reshape(-1), test_labels.reshape(-1)   \n",
+        "    model = LGBMRegressor(**params)\n",
+        "    model.fit(train_features, train_labels)\n",
+        "    preds = model.predict(WAE_x)\n",
+        "    return preds\n",
+        "\n",
+        "class Net(nn.Module):  \n",
+        "        def __init__(self, n_feature, n_hidden, n_output, w):\n",
+        "            super(Net, self).__init__()   \n",
+        "            # self.BN=torch.nn.BatchNorm1d(n_hidden)\n",
+        "            self.hidden1 = torch.nn.Linear(n_feature, n_hidden) \n",
+        "            nn.init.kaiming_normal_(self.hidden1.weight)\n",
+        "            \n",
+        "            self.hiddens = nn.ModuleList ([nn.Linear(n_hidden, n_hidden) for i in range(w)])                            \n",
+        "            for m in self.hiddens:\n",
+        "                nn.init.kaiming_normal_(m.weight)   \n",
+        "            \n",
+        "            self.predict = torch.nn.Linear(n_hidden, n_output) \n",
+        "            nn.init.kaiming_normal_(self.predict.weight)\n",
+        "    \n",
+        "        def forward(self, x): \n",
+        "            x = self.hidden1(x)\n",
+        "            # x = self.BN(x)\n",
+        "            # x = self.Dropout (x)\n",
+        "            x = F.relu(x)   \n",
+        "            \n",
+        "            for m in self.hiddens:\n",
+        "                x = m(x)\n",
+        "                # x = self.BN(x)\n",
+        "                x = F.relu(x) \n",
+        "                          \n",
+        "            x = self.predict(x)\n",
+        "            # x = self.BN_3(x)\n",
+        "            # x = self.Dropout (x)\n",
+        "            return x\n",
+        "        \n",
+        "def NN(n,seed, WAE_x):\n",
+        "    target = pd.read_excel('Invar_inference_NN.xlsx')\n",
+        "    batch_size = target.at[n,'batch_size']\n",
+        "    lr = target.at[n,'lr']\n",
+        "    module__n_hidden = target.at[n,'module__n_hidden']\n",
+        "    module__w = target.at[n,'module__w']\n",
+        "    module__n_hidden = int(module__n_hidden)\n",
+        "    module__w = int(module__w)\n",
+        "    batch_size = int(batch_size)\n",
+        "    net = Net(n_feature=17, n_hidden=module__n_hidden, n_output=1, w = module__w)\n",
+        "    print(net)\n",
+        "    #load模型\n",
+        "    net.load_state_dict(torch.load('NN_rank_{}-seed_{}.pt'.format(n,seed)))\n",
+        "    net.eval()\n",
+        "    Comp_NN = torch.FloatTensor(WAE_x.values)\n",
+        "    preds = net(Comp_NN)\n",
+        "    preds=preds.data.numpy()\n",
+        "    return preds\n",
+        "\n",
+        "#   \n",
+        "r=0\n",
+        "Comp_total = pd.DataFrame()\n",
+        "for i in range(1,3):\n",
+        "    for j in range(40,42):\n",
+        "        #Tree\n",
+        "        print ('prediction_Tree_{}'.format(r))\n",
+        "        prediction = Tree(i,j,WAE_x)\n",
+        "        Comp_total['pred_Z_Tree_{}'.format(r)] = prediction\n",
+        "        #NN\n",
+        "        print ('prediction_NN_{}'.format(r))\n",
+        "        prediction = NN(i,j,WAE_x)\n",
+        "        Comp_total['pred_Z_NN_{}'.format(r)] = prediction    \n",
+        "        r += 1"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "prediction_Tree_0\n",
+            "prediction_NN_0\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=110, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (1): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (2): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (3): Linear(in_features=110, out_features=110, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=110, out_features=1, bias=True)\n",
+            ")\n",
+            "prediction_Tree_1\n",
+            "prediction_NN_1\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=110, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (1): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (2): Linear(in_features=110, out_features=110, bias=True)\n",
+            "    (3): Linear(in_features=110, out_features=110, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=110, out_features=1, bias=True)\n",
+            ")\n",
+            "prediction_Tree_2\n",
+            "prediction_NN_2\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=229, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (1): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (2): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (3): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (4): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (5): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (6): Linear(in_features=229, out_features=229, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=229, out_features=1, bias=True)\n",
+            ")\n",
+            "prediction_Tree_3\n",
+            "prediction_NN_3\n",
+            "Net(\n",
+            "  (hidden1): Linear(in_features=17, out_features=229, bias=True)\n",
+            "  (hiddens): ModuleList(\n",
+            "    (0): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (1): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (2): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (3): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (4): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (5): Linear(in_features=229, out_features=229, bias=True)\n",
+            "    (6): Linear(in_features=229, out_features=229, bias=True)\n",
+            "  )\n",
+            "  (predict): Linear(in_features=229, out_features=1, bias=True)\n",
+            ")\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "-vPQl2BV2D6V"
+      },
+      "source": [
+        ""
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "R6hTpb8SJMFV"
+      },
+      "source": [
+        "#Rank-based exploration-exploitation strategy\n",
+        "\n",
+        "In this case, we emphasize on exploitation - more weights on mean prediction\n",
+        "(0.8) and less on std prediction(0.2). then we use the top - 6 compositions for the next stage computation (DFT-involved)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "e5oNynw2PeWQ"
+      },
+      "source": [
+        "prediciton_mean = Comp_total.mean(axis=1)\n",
+        "prediciton_std = Comp_total.std(axis=1)\n",
+        "WAE_x['prediction_mean'] = prediciton_mean\n",
+        "WAE_x['prediction_std'] = prediciton_std\n",
+        "WAE_x['rank_score'] =prediciton_std*0.2+prediciton_mean*0.8"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "PRyJ0mEvQJJv"
+      },
+      "source": [
+        "WAE = WAE_x[['Fe','Ni','Co','Cr','V','Cu','prediction_mean', 'prediction_std','rank_score']]"
+      ],
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 417
+        },
+        "id": "SBYAl3jkQ2Uu",
+        "outputId": "f40f174f-b785-4938-c31e-33c5f7036dd7"
+      },
+      "source": [
+        "WAE.sort_values(by=['rank_score'])"
+      ],
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>Fe</th>\n",
+              "      <th>Ni</th>\n",
+              "      <th>Co</th>\n",
+              "      <th>Cr</th>\n",
+              "      <th>V</th>\n",
+              "      <th>Cu</th>\n",
+              "      <th>prediction_mean</th>\n",
+              "      <th>prediction_std</th>\n",
+              "      <th>rank_score</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>2638</th>\n",
+              "      <td>0.648121</td>\n",
+              "      <td>0.351275</td>\n",
+              "      <td>0.000035</td>\n",
+              "      <td>0.000149</td>\n",
+              "      <td>0.000420</td>\n",
+              "      <td>1.327315e-06</td>\n",
+              "      <td>0.159114</td>\n",
+              "      <td>1.495680</td>\n",
+              "      <td>0.426427</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2055</th>\n",
+              "      <td>0.646740</td>\n",
+              "      <td>0.352707</td>\n",
+              "      <td>0.000029</td>\n",
+              "      <td>0.000147</td>\n",
+              "      <td>0.000376</td>\n",
+              "      <td>1.586321e-06</td>\n",
+              "      <td>0.234404</td>\n",
+              "      <td>1.482898</td>\n",
+              "      <td>0.484102</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2912</th>\n",
+              "      <td>0.622917</td>\n",
+              "      <td>0.323373</td>\n",
+              "      <td>0.051926</td>\n",
+              "      <td>0.000148</td>\n",
+              "      <td>0.001636</td>\n",
+              "      <td>1.708764e-07</td>\n",
+              "      <td>0.429156</td>\n",
+              "      <td>0.843978</td>\n",
+              "      <td>0.512121</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1930</th>\n",
+              "      <td>0.620355</td>\n",
+              "      <td>0.316188</td>\n",
+              "      <td>0.062158</td>\n",
+              "      <td>0.000187</td>\n",
+              "      <td>0.001113</td>\n",
+              "      <td>1.329685e-07</td>\n",
+              "      <td>0.455359</td>\n",
+              "      <td>0.797104</td>\n",
+              "      <td>0.523708</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>316</th>\n",
+              "      <td>0.620113</td>\n",
+              "      <td>0.316095</td>\n",
+              "      <td>0.062489</td>\n",
+              "      <td>0.000186</td>\n",
+              "      <td>0.001116</td>\n",
+              "      <td>1.330367e-07</td>\n",
+              "      <td>0.457068</td>\n",
+              "      <td>0.798083</td>\n",
+              "      <td>0.525271</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>...</th>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "      <td>...</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1130</th>\n",
+              "      <td>0.569404</td>\n",
+              "      <td>0.285598</td>\n",
+              "      <td>0.024451</td>\n",
+              "      <td>0.000002</td>\n",
+              "      <td>0.120542</td>\n",
+              "      <td>2.101832e-06</td>\n",
+              "      <td>18.341187</td>\n",
+              "      <td>3.569920</td>\n",
+              "      <td>15.386933</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1214</th>\n",
+              "      <td>0.522654</td>\n",
+              "      <td>0.282285</td>\n",
+              "      <td>0.063260</td>\n",
+              "      <td>0.000002</td>\n",
+              "      <td>0.131796</td>\n",
+              "      <td>2.618308e-06</td>\n",
+              "      <td>18.684805</td>\n",
+              "      <td>2.981920</td>\n",
+              "      <td>15.544228</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2932</th>\n",
+              "      <td>0.540877</td>\n",
+              "      <td>0.283679</td>\n",
+              "      <td>0.038131</td>\n",
+              "      <td>0.000001</td>\n",
+              "      <td>0.137310</td>\n",
+              "      <td>2.563241e-06</td>\n",
+              "      <td>18.765152</td>\n",
+              "      <td>2.666628</td>\n",
+              "      <td>15.545447</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1347</th>\n",
+              "      <td>0.505444</td>\n",
+              "      <td>0.292270</td>\n",
+              "      <td>0.058011</td>\n",
+              "      <td>0.000001</td>\n",
+              "      <td>0.144271</td>\n",
+              "      <td>3.012415e-06</td>\n",
+              "      <td>19.373516</td>\n",
+              "      <td>2.579720</td>\n",
+              "      <td>16.014757</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2725</th>\n",
+              "      <td>0.526752</td>\n",
+              "      <td>0.290878</td>\n",
+              "      <td>0.035879</td>\n",
+              "      <td>0.000001</td>\n",
+              "      <td>0.146487</td>\n",
+              "      <td>2.828880e-06</td>\n",
+              "      <td>19.460622</td>\n",
+              "      <td>2.583123</td>\n",
+              "      <td>16.085122</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "<p>2966 rows × 9 columns</p>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "            Fe        Ni        Co  ...  prediction_mean  prediction_std  rank_score\n",
+              "2638  0.648121  0.351275  0.000035  ...         0.159114        1.495680    0.426427\n",
+              "2055  0.646740  0.352707  0.000029  ...         0.234404        1.482898    0.484102\n",
+              "2912  0.622917  0.323373  0.051926  ...         0.429156        0.843978    0.512121\n",
+              "1930  0.620355  0.316188  0.062158  ...         0.455359        0.797104    0.523708\n",
+              "316   0.620113  0.316095  0.062489  ...         0.457068        0.798083    0.525271\n",
+              "...        ...       ...       ...  ...              ...             ...         ...\n",
+              "1130  0.569404  0.285598  0.024451  ...        18.341187        3.569920   15.386933\n",
+              "1214  0.522654  0.282285  0.063260  ...        18.684805        2.981920   15.544228\n",
+              "2932  0.540877  0.283679  0.038131  ...        18.765152        2.666628   15.545447\n",
+              "1347  0.505444  0.292270  0.058011  ...        19.373516        2.579720   16.014757\n",
+              "2725  0.526752  0.290878  0.035879  ...        19.460622        2.583123   16.085122\n",
+              "\n",
+              "[2966 rows x 9 columns]"
+            ]
+          },
+          "metadata": {},
+          "execution_count": 60
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "Z9nTu3ov2E5d"
+      },
+      "source": [
+        "Congrats! you have completed the tutorial!"
+      ]
+    }
+  ]
+}
\ No newline at end of file