diff --git a/assets/kaggle_competition/Logo_MPG.png b/assets/kaggle_competition/logo_MPG.png
similarity index 100%
rename from assets/kaggle_competition/Logo_MPG.png
rename to assets/kaggle_competition/logo_MPG.png
diff --git a/assets/kaggle_competition/Logo_NOMAD.png b/assets/kaggle_competition/logo_NOMAD.png
similarity index 100%
rename from assets/kaggle_competition/Logo_NOMAD.png
rename to assets/kaggle_competition/logo_NOMAD.png
diff --git a/kaggle_competition.ipynb b/kaggle_competition.ipynb
index 6d989f30f5cad04c9505aebf3bb03c232add1da3..88c61cb20f80c2d7f468df814163fbdc5ce9a27f 100644
--- a/kaggle_competition.ipynb
+++ b/kaggle_competition.ipynb
@@ -2,11 +2,11 @@
"cells": [
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 1,
"metadata": {
"ExecuteTime": {
- "end_time": "2021-01-19T10:23:31.012040Z",
- "start_time": "2021-01-19T10:23:31.009452Z"
+ "end_time": "2021-11-11T17:10:52.300325Z",
+ "start_time": "2021-11-11T17:10:52.290470Z"
},
"init_cell": true
},
@@ -29,44 +29,15 @@
}
},
"source": [
- "<div id=\"teaser\" style=' background-position: right center; background-size: 00px; background-repeat: no-repeat; \n",
- " padding-top: 20px;\n",
- " padding-right: 10px;\n",
- " padding-bottom: 170px;\n",
- " padding-left: 10px;\n",
- " border-bottom: 14px double #333;\n",
- " border-top: 14px double #333;' > \n",
- "\n",
- " \n",
- " <div style=\"text-align:center\">\n",
- " <b><font size=\"6.4\">2018 NOMAD-Kaggle research competition</font></b> \n",
- " </div>\n",
- " \n",
- "<p>\n",
- " created by:\n",
- " Xiangyue Liu<sup>1</sup> (<a href=\"mailto:xyliu@fhi-berlin.mpg.de\">email</a>),\n",
- " Christopher Sutton<sup>1</sup> (<a href=\"mailto:sutton@fhi-berlin.mpg.de\">email</a>),\n",
- " Luca M. Ghiringhelli<sup>1</sup>(<a href=\"mailto:ghiringhelli@fhi-berlin.mpg.de\">email</a>),\n",
- " Takenori Yamamoto<sup>2</sup>, \n",
- " Lars Blumenthal<sup>3,4</sup>,\n",
- " Jacek Golebiowski<sup>3,4</sup>, \n",
- " Angelo Ziletti<sup>1</sup>, \n",
- " and Matthias Scheffler<sup>1</sup>\n",
- " \n",
- "<sup>1</sup> Fritz Haber Institute of the Max Planck Society, Faradayweg 4-6, D-14195 Berlin, Germany <br>\n",
- "<sup>2</sup> Research Institute for Mathematical and Computational Sciences (RIMCS), LLC, Yokohama, Japan <br>\n",
- "<sup>4</sup> EPSRC Centre for Doctoral Training on Theory and Simulation of Materials Department of Physics, Imperial College London, London, U.K. <br>\n",
- "<sup>5</sup> Thomas Young Centre for Theory and Simulation of Materials, Department of Materials, Imperial College London, London, U.K <br>\n",
- " \n",
- " \n",
- " \n",
- "<span class=\"nomad--last-updated\" data-version=\"v1.0.0\">[Last updated: January 19, 2021]</span>\n",
- " \n",
- "<div> \n",
- "<img style=\"float: left;\" src=\"assets/kaggle_competition/Logo_MPG.png\" width=\"200\"> \n",
- "<img style=\"float: right;\" src=\"assets/kaggle_competition/Logo_NOMAD.png\" width=\"250\">\n",
- "</div>\n",
- "</div>\n"
+ "<img src=\"assets/kaggle_competition/header.jpg\" width=\"900\"> "
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "<img style=\"float: left;\" src=\"assets/kaggle_competition/logo_NOMAD.png\" width=300>\n",
+ "<img style=\"float: right;\" src=\"assets/kaggle_competition/logo_MPG.png\" width=170> "
]
},
{
@@ -106,16 +77,45 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 2,
"metadata": {
"ExecuteTime": {
- "end_time": "2021-01-19T10:23:31.032317Z",
- "start_time": "2021-01-19T10:23:31.014029Z"
+ "end_time": "2021-11-11T17:10:52.313149Z",
+ "start_time": "2021-11-11T17:10:52.302093Z"
},
"init_cell": true,
"scrolled": false
},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "<script>\n",
+ " code_show=true; \n",
+ " function code_toggle() {\n",
+ " if (code_show)\n",
+ " {\n",
+ " $('div.input').hide();\n",
+ " } \n",
+ " else \n",
+ " {\n",
+ " $('div.input').show();\n",
+ " }\n",
+ " code_show = !code_show\n",
+ " } \n",
+ " $( document ).ready(code_toggle);\n",
+ "</script>\n",
+ "The Python code for this notebook is by default hidden for easier reading.\n",
+ "To toggle on/off the raw code, click <a href=\"javascript:code_toggle()\">here</a>.\n"
+ ],
+ "text/plain": [
+ "<IPython.core.display.HTML object>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
"%%HTML\n",
"<script>\n",
@@ -150,16 +150,232 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 3,
"metadata": {
"ExecuteTime": {
- "end_time": "2021-01-19T10:23:31.049437Z",
- "start_time": "2021-01-19T10:23:31.034656Z"
+ "end_time": "2021-11-11T17:10:52.319760Z",
+ "start_time": "2021-11-11T17:10:52.314773Z"
},
"init_cell": true,
"scrolled": false
},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "<!-- CSS Style Inline: -->\n",
+ " <style type=\"text/css\">\n",
+ " #jmol_div227{\n",
+ " height: 350px;\n",
+ " width: 350px;\n",
+ " float: left;\n",
+ " }\n",
+ " #jmol_div12{\n",
+ " height: 350px;\n",
+ " width: 350px;\n",
+ " float: left;\n",
+ " }\n",
+ " #jmol_div206{\n",
+ " height: 350px;\n",
+ " width: 350px;\n",
+ " float: left;\n",
+ " }\n",
+ " #jmol_div33{\n",
+ " height: 350px;\n",
+ " width: 350px;\n",
+ " float: left;\n",
+ " }\n",
+ " #jmol_div194{\n",
+ " height: 350px;\n",
+ " width: 350px;\n",
+ " float: left;\n",
+ " }\n",
+ " #jmol_div167{\n",
+ " height: 350px;\n",
+ " width: 350px;\n",
+ " float: left;\n",
+ " }\n",
+ " </style>\n",
+ "\n",
+ " <!-- Load Jmol javascript library -->\n",
+ " <script type=\"text/javascript\" src=\"assets/kaggle_competition/jsmol/JSmol.min.js\"></script>\n",
+ "\n",
+ " <!-- calls to jQuery and Jmol (inline) -->\n",
+ " <script type=\"text/javascript\">\n",
+ " // Jmol readyFunction \n",
+ " jmol_isReady = function(applet) {\n",
+ " document.title = (applet._id + \" - Jmol \" + Jmol.___JmolVersion)\n",
+ " Jmol._getElement(applet, \"appletdiv\").style.border=\"0px solid blue\"\n",
+ " }\n",
+ " // initialize Jmol Applet\n",
+ " var myJmol227 = \"myJmol227\";\n",
+ " var Info227 = {\n",
+ " width: \"100%\",\n",
+ " height: \"100%\",\n",
+ " color: \"#ffffff\", \n",
+ " use: \"HTML5\",\n",
+ " j2sPath: \"assets/kaggle_competition/jsmol/j2s\",\n",
+ " jarPath: \"assets/kaggle_competition/jsmol/java\",\n",
+ " jarFile: \"JmolAppletSigned.jar\",\n",
+ " debug: false,\n",
+ " readyFunction: jmol_isReady,\n",
+ " script: 'load \"data/kaggle_competition/xyz/geometry3-spacegroup227.xyz\" ;',\n",
+ " allowJavaScript: false,\n",
+ " disableJ2SLoadMonitor: true,\n",
+ " }\n",
+ " var myJmol12 = \"myJmol12\";\n",
+ " var Info12 = {\n",
+ " width: \"100%\",\n",
+ " height: \"100%\",\n",
+ " color: \"#ffffff\", \n",
+ " use: \"HTML5\",\n",
+ " j2sPath: \"assets/kaggle_competition/jsmol/j2s\",\n",
+ " jarPath: \"assets/kaggle_competition/jsmol/java\",\n",
+ " jarFile: \"JmolAppletSigned.jar\",\n",
+ " debug: false,\n",
+ " readyFunction: jmol_isReady,\n",
+ " script: 'load \"data/kaggle_competition/xyz/geometry8-spacegroup12.xyz\" ;',\n",
+ " allowJavaScript: false,\n",
+ " disableJ2SLoadMonitor: true,\n",
+ " }\n",
+ " var myJmol206 = \"myJmol206\";\n",
+ " var Info206 = {\n",
+ " width: \"100%\",\n",
+ " height: \"100%\",\n",
+ " color: \"#ffffff\", \n",
+ " use: \"HTML5\",\n",
+ " j2sPath: \"assets/kaggle_competition/jsmol/j2s\",\n",
+ " jarPath: \"assets/kaggle_competition/jsmol/java\",\n",
+ " jarFile: \"JmolAppletSigned.jar\",\n",
+ " debug: false,\n",
+ " readyFunction: jmol_isReady,\n",
+ " script: 'load \"data/kaggle_competition/xyz/geometry7-spacegroup206.xyz\" ;',\n",
+ " allowJavaScript: false,\n",
+ " disableJ2SLoadMonitor: true,\n",
+ " }\n",
+ " var myJmol33 = \"myJmol33\";\n",
+ " var Info33 = {\n",
+ " width: \"100%\",\n",
+ " height: \"100%\",\n",
+ " color: \"#ffffff\", \n",
+ " use: \"HTML5\",\n",
+ " j2sPath: \"assets/kaggle_competition/jsmol/j2s\",\n",
+ " jarPath: \"assets/kaggle_competition/jsmol/java\",\n",
+ " jarFile: \"JmolAppletSigned.jar\",\n",
+ " debug: false,\n",
+ " readyFunction: jmol_isReady,\n",
+ " script: 'load \"data/kaggle_competition/xyz/geometry1-spacegroup33.xyz\" ;',\n",
+ " allowJavaScript: false,\n",
+ " disableJ2SLoadMonitor: true,\n",
+ " }\n",
+ " var myJmol194 = \"myJmol194\";\n",
+ " var Info194 = {\n",
+ " width: \"100%\",\n",
+ " height: \"100%\",\n",
+ " color: \"#ffffff\", \n",
+ " use: \"HTML5\",\n",
+ " j2sPath: \"assets/kaggle_competition/jsmol/j2s\",\n",
+ " jarPath: \"assets/kaggle_competition/jsmol/java\",\n",
+ " jarFile: \"JmolAppletSigned.jar\",\n",
+ " debug: false,\n",
+ " readyFunction: jmol_isReady,\n",
+ " script: 'load \"data/kaggle_competition/xyz/geometry2-spacegroup194.xyz\" ;',\n",
+ " allowJavaScript: false,\n",
+ " disableJ2SLoadMonitor: true,\n",
+ " }\n",
+ " var myJmol167 = \"myJmol167\";\n",
+ " var Info167 = {\n",
+ " width: \"100%\",\n",
+ " height: \"100%\",\n",
+ " color: \"#ffffff\", \n",
+ " use: \"HTML5\",\n",
+ " j2sPath: \"assets/kaggle_competition/jsmol/j2s\",\n",
+ " jarPath: \"assets/kaggle_competition/jsmol/java\",\n",
+ " jarFile: \"JmolAppletSigned.jar\",\n",
+ " debug: false,\n",
+ " readyFunction: jmol_isReady,\n",
+ " script: 'load \"data/kaggle_competition/xyz/geometry4-spacegroup167.xyz\" ;',\n",
+ " allowJavaScript: false,\n",
+ " disableJ2SLoadMonitor: true,\n",
+ " }\n",
+ " // jQuery ready functions\n",
+ " // is called when page has been completely loaded\n",
+ " $(document).ready(function() {\n",
+ " $(\"#jmol_div227\").html(Jmol.getAppletHtml(myJmol227, Info227))\n",
+ " })\n",
+ " $(document).ready(function() {\n",
+ " $(\"#jmol_div12\").html(Jmol.getAppletHtml(myJmol12, Info12))\n",
+ " })\n",
+ " $(document).ready(function() {\n",
+ " $(\"#jmol_div206\").html(Jmol.getAppletHtml(myJmol206, Info206))\n",
+ " })\n",
+ " $(document).ready(function() {\n",
+ " $(\"#jmol_div33\").html(Jmol.getAppletHtml(myJmol33, Info33))\n",
+ " })\n",
+ " $(document).ready(function() {\n",
+ " $(\"#jmol_div194\").html(Jmol.getAppletHtml(myJmol194, Info194))\n",
+ " })\n",
+ " $(document).ready(function() {\n",
+ " $(\"#jmol_div167\").html(Jmol.getAppletHtml(myJmol167, Info167))\n",
+ " })\n",
+ " var lastPrompt=0;\n",
+ " \n",
+ " \n",
+ " \n",
+ " function show_hide_structures()\n",
+ " {\n",
+ " var x = document.getElementById(\"geometry_jmol\");\n",
+ " if (x.style.display === \"none\") \n",
+ " {\n",
+ " x.style.display = \"block\";\n",
+ " } \n",
+ " else \n",
+ " {\n",
+ " x.style.display = \"none\";\n",
+ " }\n",
+ " }\n",
+ " \n",
+ " </script>\n",
+ "\n",
+ "<div>\n",
+ " <button type=\"button\" style=\"background-color:#f2f2f2;border:#555555;border-radius: 4px;font-size: 16px; width:200px; height:30px\" onclick=\"show_hide_structures()\">Show/hide structures</button>\n",
+ " <br><br>\n",
+ "</div>\n",
+ " \n",
+ "<div id=\"geometry_jmol\">\n",
+ " <table>\n",
+ " <tr>\n",
+ " <th><center>Spacegroup 12 (C2/m)</center></th>\n",
+ " <th><center>Spacegroup 33 (Pna2<sub>1</sub>)</center></th>\n",
+ " <th><center>Spacegroup 167 (R<font style=\"text-decoration: overline\">3</font>c)</center></th>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th><div id='jmol_div12'></div></th>\n",
+ " <th><div id='jmol_div33'></div></th>\n",
+ " <th><div id='jmol_div167'></div></th>\n",
+ " </tr>\n",
+ " \n",
+ " <tr>\n",
+ " <th><center>Spacegroup 194 (P6<sub>3</sub>/mmc)</center></th>\n",
+ " <th><center>Spacegroup 206 (Ia<font style=\"text-decoration: overline\">3</font>)</center></th>\n",
+ " <th><center>Spacegroup 227 (Fd<font style=\"text-decoration: overline\">3</font>m)</center></th>\n",
+ " </tr>\n",
+ " <tr>\n",
+ " <th><div id='jmol_div194'></div></th>\n",
+ " <th><div id='jmol_div206'></div></th>\n",
+ " <th><div id='jmol_div227'></div></th>\n",
+ " </tr>\n",
+ " </table>\n",
+ "</div>\n"
+ ],
+ "text/plain": [
+ "<IPython.core.display.HTML object>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
"%%html\n",
"<!-- CSS Style Inline: -->\n",
@@ -434,16 +650,466 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 4,
"metadata": {
"ExecuteTime": {
- "end_time": "2021-01-19T10:23:31.064353Z",
- "start_time": "2021-01-19T10:23:31.051499Z"
+ "end_time": "2021-11-11T17:10:52.329322Z",
+ "start_time": "2021-11-11T17:10:52.322187Z"
},
"init_cell": true,
"scrolled": false
},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "<br><br><br>\n",
+ "<font size=\"6.5em\"><b>Make predictions on formation energies and bandgaps</b></font>\n",
+ "<br><hr><br>\n",
+ "<font size=\"5em\"><b>Winning representations combined with different regression methods</b></font>\n",
+ "<br><br>\n",
+ "<font size = \"3.5em\"> To understand the relative importance of the representation vs. regression model, one can examine the performance of each representation combined with different regression models. \n",
+ "The hyperparameters are optimized for each representation/regressor combination. </font>\n",
+ "<br>\n",
+ "<font size = \"3.5em\" color=\"009FC2\"><br>Warning: the learning algorithm employed in this study (e.g. grid-search) can not guarantee deterministic results. The actual predictions can divergent from the published data.\n",
+ "</font>\n",
+ "\n",
+ "<br><br><br>\n",
+ "\n",
+ "<form>\n",
+ " <font size=\"4em\">Select a representation and a regression method:</font>\n",
+ " <select id=\"representation\" onclick=\"show_options()\">\n",
+ " <option value=\"ngram\">n-gram</option>\n",
+ " <option value=\"soap\">SOAP</option>\n",
+ " </select>\n",
+ " <select id=\"regression\" onclick=\"show_options()\">\n",
+ " <option value=\"krr\">KRR</option>\n",
+ " <option value=\"nn\">Neural network</option>\n",
+ " </select>\n",
+ " \n",
+ " <button type=\"button\" id=\"button_show_options\" style=\"background-color:#f2f2f2;border:#555555;border-radius: 4px;font-size: 16px; width:150px; height:30px\" onclick=\"show_options()\">More options</button>\n",
+ " <button type=\"button\" id=\"button_hide_options\" style=\"background-color:#f2f2f2;border:#555555;border-radius: 4px;font-size: 16px; width:150px; height:30px; display: none\" onclick=\"hide_options()\">Less options</button>\n",
+ "\n",
+ "</form>\n",
+ "\n",
+ "\n",
+ "\n",
+ "<script type=\"text/Javascript\">\n",
+ "\n",
+ " function show_options()\n",
+ " {\n",
+ " representation_value = document.getElementById(\"representation\").value;\n",
+ " regression_value = document.getElementById(\"regression\").value;\n",
+ " \n",
+ " if(representation_value == \"ngram\")\n",
+ " {\n",
+ " document.getElementById(\"div_options_soap\").style.display=\"none\";\n",
+ " document.getElementById(\"div_options_ngram\").style.display=\"\";\n",
+ " document.getElementById(\"button_show_options\").style.display=\"none\";\n",
+ " document.getElementById(\"button_hide_options\").style.display=\"\"; \n",
+ " }\n",
+ " else if(representation_value == \"soap\")\n",
+ " {\n",
+ " document.getElementById(\"div_options_ngram\").style.display=\"none\";\n",
+ " document.getElementById(\"div_options_soap\").style.display=\"\";\n",
+ " document.getElementById(\"button_show_options\").style.display=\"none\";\n",
+ " document.getElementById(\"button_hide_options\").style.display=\"\"; \n",
+ " }\n",
+ " if(regression_value == \"nn\")\n",
+ " {\n",
+ " document.getElementById(\"div_options_nn\").style.display=\"\"; \n",
+ " set_nn_default();\n",
+ " }\n",
+ " else if(regression_value == \"krr\")\n",
+ " {\n",
+ " document.getElementById(\"div_options_nn\").style.display=\"none\"; \n",
+ " }\n",
+ " }\n",
+ " function hide_options()\n",
+ " {\n",
+ " document.getElementById(\"button_show_options\").style.display=\"\";\n",
+ " document.getElementById(\"button_hide_options\").style.display=\"none\";\n",
+ " document.getElementById(\"div_options_ngram\").style.display=\"none\";\n",
+ " document.getElementById(\"div_options_soap\").style.display=\"none\";\n",
+ " document.getElementById(\"div_options_nn\").style.display=\"none\";\n",
+ " }\n",
+ " function set_nn_n_neurons()\n",
+ " {\n",
+ " document.getElementById(\"div_options_nn_n_neurons\").style.display=\"\";\n",
+ " n_layers = document.getElementById(\"input_n_nn\").value;\n",
+ " \n",
+ " contents = \"<br><font size=\\\"3em\\\">Number of neurons in each hidden layer</font><br>\"\n",
+ " for(i=0; i<n_layers; i++)\n",
+ " {\n",
+ " content_add = \"<font size=\\\"3em\\\"> Layer \" + String(i+1) + \": </font><input\\ type=\\\"number\\\"\\ id=\\\"input_n_neurons_\" + String(i) + \"\\\"\\ value=\\\"256\\\"\\ min=\\\"1\\\"\\ max=\\\"1024\\\"\\ >\";\n",
+ " contents += content_add;\n",
+ " if((i+1)%6 ==0)\n",
+ " {\n",
+ " contents += \"<br>\"\n",
+ " }\n",
+ " }\n",
+ " //alert(contents);\n",
+ " document.getElementById(\"div_options_nn_n_neurons\").innerHTML = contents;\n",
+ " }\n",
+ " function set_nn_default()\n",
+ " {\n",
+ " document.getElementById(\"div_options_nn_n_neurons\").style.display=\"\";\n",
+ " representation_value = document.getElementById(\"representation\").value;\n",
+ " if(representation_value == \"ngram\")\n",
+ " {\n",
+ " n_layers = 11;\n",
+ " document.getElementById(\"input_n_nn\").value = n_layers;\n",
+ " contents = \"<br><font size=\\\"3em\\\">Number of neurons in each hidden layer</font><br>\"\n",
+ " for(i=0; i<7; i++)\n",
+ " {\n",
+ " content_add = \"<font size=\\\"3em\\\"> Layer \" + String(i+1) + \": </font><input\\ type=\\\"number\\\"\\ id=\\\"input_n_neurons_\" + String(i) + \"\\\"\\ value=\\\"100\\\"\\ min=\\\"1\\\"\\ max=\\\"1024\\\"\\ >\";\n",
+ " contents += content_add;\n",
+ " if((i+1)%6 ==0)\n",
+ " {\n",
+ " contents += \"<br>\"\n",
+ " }\n",
+ "\n",
+ " }\n",
+ " for(i=7; i<11; i++)\n",
+ " {\n",
+ " content_add = \"<font size=\\\"3em\\\"> Layer \" + String(i+1) + \": </font><input\\ type=\\\"number\\\"\\ id=\\\"input_n_neurons_\" + String(i) + \"\\\"\\ value=\\\"50\\\"\\ min=\\\"1\\\"\\ max=\\\"1024\\\"\\ >\";\n",
+ " contents += content_add;\n",
+ " if((i+1)%6 ==0)\n",
+ " {\n",
+ " contents += \"<br>\"\n",
+ " }\n",
+ " }\n",
+ " }\n",
+ " else if(representation_value == \"soap\")\n",
+ " {\n",
+ " n_layers = 3;\n",
+ " document.getElementById(\"input_n_nn\").value = n_layers;\n",
+ " \n",
+ " contents = \"<br><font size=\\\"3em\\\">Number of neurons in each hidden layer</font><br>\"\n",
+ " for(i=0; i<1; i++)\n",
+ " {\n",
+ " content_add = \"<font size=\\\"3em\\\"> Layer \" + String(i+1) + \": </font><input\\ type=\\\"number\\\"\\ id=\\\"input_n_neurons_\" + String(i) + \"\\\"\\ value=\\\"512\\\"\\ min=\\\"1\\\"\\ max=\\\"1024\\\"\\ >\";\n",
+ " contents += content_add;\n",
+ " }\n",
+ " for(i=1; i<3; i++)\n",
+ " {\n",
+ " content_add = \"<font size=\\\"3em\\\"> Layer \" + String(i+1) + \": </font><input\\ type=\\\"number\\\"\\ id=\\\"input_n_neurons_\" + String(i) + \"\\\"\\ value=\\\"256\\\"\\ min=\\\"1\\\"\\ max=\\\"1024\\\"\\ >\";\n",
+ " contents += content_add;\n",
+ " }\n",
+ " }\n",
+ " \n",
+ " document.getElementById(\"div_options_nn_n_neurons\").innerHTML = contents;\n",
+ " save_options_nn();\n",
+ " \n",
+ " }\n",
+ " function save_options_nn()\n",
+ " {\n",
+ " n_layers = document.getElementById(\"input_n_nn\").value;\n",
+ " var n_neurons = [];\n",
+ " for(i=0; i<n_layers; i++)\n",
+ " {\n",
+ " id_i = \"input_n_neurons_\" + String(i);\n",
+ " n_neurons_i = document.getElementById(id_i).value;\n",
+ " n_neurons.push(n_neurons_i);\n",
+ " }\n",
+ " var command = \"N_nn_neurons = \" + n_neurons + \";\";\n",
+ " command += \"N_nn_layers = int(\" + n_layers + \");\"\n",
+ " var kernel = IPython.notebook.kernel;\n",
+ " kernel.execute(command);\n",
+ " \n",
+ " }\n",
+ " function save_options()\n",
+ " {\n",
+ " representation_value = document.getElementById(\"representation\").value;\n",
+ " regression_value = document.getElementById(\"regression\").value;\n",
+ " \n",
+ " if(representation_value == \"ngram\")\n",
+ " {\n",
+ " ngram_N = document.getElementById(\"input_ngram_N\").value;\n",
+ " n_threads = document.getElementById(\"input_n_threads\").value;\n",
+ " var command = \"ngram_N = int(\" + ngram_N + \"); n_threads = int(\" + n_threads + \");\" \n",
+ " }\n",
+ " if(representation_value == \"soap\")\n",
+ " {\n",
+ " soap_cutoff = document.getElementById(\"input_soap_cutoff\").value;\n",
+ " soap_lmax = document.getElementById(\"input_soap_lmax\").value;\n",
+ " soap_nmax = document.getElementById(\"input_soap_nmax\").value;\n",
+ " n_threads = document.getElementById(\"input_n_threads\").value;\n",
+ " var command = \"soap_cutoff = int(\" + soap_cutoff + \");\";\n",
+ " command += \"soap_lmax = int(\" + soap_lmax + \");\"; \n",
+ " command += \"soap_nmax = int(\" + soap_nmax + \");\";\n",
+ " command += \"n_threads = int(\" + n_threads + \");\";\n",
+ " }\n",
+ " \n",
+ " console.log(\"Executing Command: \" + command); \n",
+ " var kernel = IPython.notebook.kernel;\n",
+ " kernel.execute(command);\n",
+ " }\n",
+ "\n",
+ "</script>\n",
+ "\n",
+ "<div id=\"div_options_ngram\" style=\"display: none; position:relative; left:5%\">\n",
+ " <br>\n",
+ " <font size=\"3em\">N-grams size (N = 1 ~ 4)</font> <input type=\"number\" id=\"input_ngram_N\" value=\"3\" min=\"1\" max=\"4\" >\n",
+ " <br>\n",
+ " <font size=\"3em\">Number of threads used in regression (N = 1 ~ 4)</font> <input type=\"number\" id=\"input_n_threads\" value=\"4\" min=\"1\" max=\"4\" >\n",
+ " <br>\n",
+ " <button type=\"button\" id=\"save_options_ngram\" style=\"background-color:#f2f2f2;border:#555555;border-radius: 4px;font-size: 16px; width:150px; height:30px;\" onclick=\"save_options()\">Save options</button>\n",
+ "</div>\n",
+ "\n",
+ "<div id=\"div_options_soap\" style=\"display: none; position:relative; left:5%\">\n",
+ " <br>\n",
+ " <font size=\"3em\">SOAP cutoff (Ang) </font> <input type=\"number\" id=\"input_soap_cutoff\" value=\"10\" min=\"1\" max=\"25\">\n",
+ " <font size=\"3em\">Max. l </font><input type=\"number\" id=\"input_soap_lmax\" value=\"4\" min=\"1\" max=\"8\" >\n",
+ " <font size=\"3em\">Max. n </font><input type=\"number\" id=\"input_soap_nmax\" value=\"4\" min=\"1\" max=\"8\" >\n",
+ " <br>\n",
+ " <font size=\"3em\">Number of threads used in regression (N = 1 ~ 4)</font> <input type=\"number\" id=\"input_n_threads\" value=\"4\" min=\"1\" max=\"4\" >\n",
+ " <br>\n",
+ " <button type=\"button\" id=\"save_options_soap\" style=\"background-color:#f2f2f2;border:#555555;border-radius: 4px;font-size: 16px; width:150px; height:30px;\" onclick=\"save_options()\">Save options</button>\n",
+ "</div>\n",
+ "<br>\n",
+ "<div id=\"div_options_nn\" style=\"display: none; position:relative; left:5%\">\n",
+ " <br>\n",
+ " <font size=\"3em\">Number of (linear) hidden layers </font> <input type=\"number\" id=\"input_n_nn\" value=\"2\" min=\"1\" max=\"20\">\n",
+ " <button type=\"button\" id=\"set_n_neurons\" style=\"background-color:#f2f2f2;border:#555555;border-radius: 4px;font-size: 16px; width:250px; height:30px;\" onclick=\"set_nn_n_neurons()\">Set the number of neurons</button>\n",
+ " \n",
+ " <div id=\"div_options_nn_n_neurons\" style=\"display: none; position:relative; left:0%\">\n",
+ " <br>\n",
+ " <font size=\"3em\">Number of neurons in each hidden layer</font>\n",
+ " </div>\n",
+ " <br>\n",
+ " <button type=\"button\" id=\"set_nn_default\" style=\"background-color:#f2f2f2;border:#555555;border-radius: 4px;font-size: 16px; width:200px; height:30px;\" onclick=\"set_nn_default()\">Use default configuration</button>\n",
+ " <button type=\"button\" id=\"save_options_nn\" style=\"background-color:#f2f2f2;border:#555555;border-radius: 4px;font-size: 16px; width:150px; height:30px;\" onclick=\"save_options_nn()\">Save options</button>\n",
+ "</div>\n",
+ "\n",
+ "<div id=\"demoa\"></div>\n",
+ "<br><br>\n",
+ "\n",
+ "<button type=\"button\" id=\"button_show_sample_output\" style=\"background-color:#f2f2f2;border:#555555;border-radius: 4px;font-size: 16px; width:200px; height:30px\" onclick=\"show_sample_output()\">Show sample outputs</button>\n",
+ "<button type=\"button\" id=\"button_hide_sample_output\" style=\"background-color:#f2f2f2;border:#555555;border-radius: 4px;font-size: 16px; width:200px; height:30px; display: none\" onclick=\"hide_sample_output()\">Hide sample outputs</button>\n",
+ "\n",
+ "<button type=\"button\" style=\"background-color:#f2f2f2;border:#555555;border-radius: 4px;font-size: 16px; width:150px; height:30px\" onclick=\"get_repr_regr_combination()\">Get predictions</button>\n",
+ "<input type=\"checkbox\" id=\"if_learning_curve\" value=\"learning_curve\" onclick=show_warning_learning_curve()> Show learning curve\n",
+ "\n",
+ "<input type=\"checkbox\" id=\"if_use_prestored\" value=\"use_prestored\" onclick=show_warning_use_prestored_model() checked=true> Use prestored models\n",
+ "<div id=\"warning_learning_curve\" style=\"display: none\">\n",
+ " <br>\n",
+ " <font color=\"009FC2\">\n",
+ " Warning: It can be very time-consuming (5~20 min/point, depending on method, model size, and number of threads employed) when getting a learning curve.\n",
+ " </font>\n",
+ " <p><font size=\"3em\">Number of points in learning curve: <input type=\"number\" id=\"N_learning_curve\" value=\"4\" min=\"1\" max=\"25\" style=\"display:none\"></font></p>\n",
+ "</div>\n",
+ "<div id=\"warning_use_prestored\" style=\"display: none\">\n",
+ " <br>\n",
+ " <font color=\"009FC2\">\n",
+ " Warning: It can be very time-consuming (10~20 min, depending on method, model size, and number of threads employed) to generate models on the fly.\n",
+ " </font>\n",
+ "</div>\n",
+ "\n",
+ "\n",
+ "<div id=\"sample_output\" style=\"display: none\">\n",
+ " <br><br>\n",
+ " <font size=\"4em\"><b> Sample outputs of predictions on formation energy and bandgap</b></font>\n",
+ " <br><br>\n",
+ " <font size=\"3em\">Representation: N-grams<br><br>Regression method: Kernel-ridge regression (KRR)</font>\n",
+ " <br><br>\n",
+ " <img src=\"assets/kaggle_competition/results-ngram-krr.png\" width=\"60%\">\n",
+ " <br><br>\n",
+ " \n",
+ "</div>\n",
+ "\n",
+ "<style>\n",
+ "* {\n",
+ " box-sizing: border-box;\n",
+ "}\n",
+ "\n",
+ ".column {\n",
+ " float: left;\n",
+ " width: 33.33%;\n",
+ " padding: 5px;\n",
+ "}\n",
+ "\n",
+ "/* Clearfix (clear floats) */\n",
+ ".row::after {\n",
+ " content: \"\";\n",
+ " clear: both;\n",
+ " display: table;\n",
+ "}\n",
+ "</style>\n",
+ "\n",
+ "<div id=\"sample_learning_curve\" style=\"display: none\">\n",
+ " <font size=\"4em\"><b> Sample outputs of learning curves for formation energy and bandgap predictions</b></font>\n",
+ " <br><br>\n",
+ " <font size=\"3em\">Representation: N-gram<br><br>Regression method: Kernel-ridge regression (KRR)</font>\n",
+ " <br><br>\n",
+ " <div style=\"display: table; width: 95%\"> \n",
+ " <div class=\"row\">\n",
+ " <div class=\"column\">\n",
+ " <img src=\"./imgs/learning_curve-formation-ngram-krr.png\" style=\"width:100%\">\n",
+ " </div>\n",
+ " <div class=\"column\">\n",
+ " <img src=\"./imgs/learning_curve-bandgap-ngram-krr.png\" style=\"width:100%\">\n",
+ " </div>\n",
+ " </div>\n",
+ " </div>\n",
+ " <br><br>\n",
+ "</div>\n",
+ "\n",
+ "<script type=\"text/Javascript\">\n",
+ "\n",
+ " window.findCellIndicesByTag = function findCellIndicesByTag(tagName) {\n",
+ " return (Jupyter.notebook.get_cells()\n",
+ " .filter(\n",
+ " ({metadata: {tags}}) => tags && tags.includes(tagName)\n",
+ " )\n",
+ " .map((cell) => Jupyter.notebook.find_cell_index(cell))\n",
+ " );\n",
+ " };\n",
+ "\n",
+ "\n",
+ " window.runCells = function runPlotCells(tags) {\n",
+ " var c = window.findCellIndicesByTag(tags);\n",
+ " Jupyter.notebook.execute_cells(c);\n",
+ " };\n",
+ " \n",
+ " function show_sample_output()\n",
+ " {\n",
+ " document.getElementById(\"button_show_sample_output\").style.display=\"none\";\n",
+ " document.getElementById(\"button_hide_sample_output\").style.display=\"\";\n",
+ " \n",
+ " var check_learning_curve = document.getElementById(\"if_learning_curve\");\n",
+ " var learning_curve_output = document.getElementById(\"sample_learning_curve\");\n",
+ " var output = document.getElementById(\"sample_output\");\n",
+ " //var representation_value = document.getElementById(\"representation\").value;\n",
+ " //var regression_value = document.getElementById(\"regression\").value;\n",
+ " \n",
+ " output.style.display = \"block\";\n",
+ " \n",
+ " \n",
+ " \n",
+ " \n",
+ " if (check_learning_curve.checked == true)\n",
+ " {\n",
+ " learning_curve_output.style.display = \"block\"; \n",
+ " } \n",
+ " else \n",
+ " {\n",
+ " learning_curve_output.style.display = \"none\";\n",
+ " }\n",
+ " }\n",
+ "\n",
+ " function hide_sample_output()\n",
+ " {\n",
+ " document.getElementById(\"button_show_sample_output\").style.display=\"\";\n",
+ " document.getElementById(\"button_hide_sample_output\").style.display=\"none\";\n",
+ "\n",
+ " var learning_curve_output = document.getElementById(\"sample_learning_curve\");\n",
+ " var output = document.getElementById(\"sample_output\");\n",
+ "\n",
+ " output.style.display = \"none\"\n",
+ " learning_curve_output.style.display = \"none\";\n",
+ " }\n",
+ " \n",
+ " function show_warning_learning_curve()\n",
+ " {\n",
+ " var checkBox = document.getElementById(\"if_learning_curve\");\n",
+ " learning_curve_output = document.getElementById(\"sample_learning_curve\");\n",
+ " var warning = document.getElementById(\"warning_learning_curve\");\n",
+ " if (checkBox.checked == true)\n",
+ " {\n",
+ " warning.style.display = \"block\";\n",
+ " document.getElementById(\"N_learning_curve\").style.display=\"\";\n",
+ " var N_learning_curve = document.getElementById(\"N_learning_curve\").value;\n",
+ " var command = \" N_learning_curve = int(\" + N_learning_curve + \");\";\n",
+ " var kernel = IPython.notebook.kernel;\n",
+ " kernel.execute(command);\n",
+ " \n",
+ " } \n",
+ " else \n",
+ " {\n",
+ " warning.style.display = \"none\";\n",
+ " document.getElementById(\"N_learning_curve\").style.display=\"none\";\n",
+ " }\n",
+ " \n",
+ " }\n",
+ " \n",
+ " function show_warning_use_prestored_model()\n",
+ " {\n",
+ " var checkBox = document.getElementById(\"if_use_prestored\");\n",
+ " var warning = document.getElementById(\"warning_use_prestored\");\n",
+ " if (checkBox.checked == true)\n",
+ " {\n",
+ " warning.style.display = \"none\";\n",
+ " \n",
+ " } \n",
+ " else \n",
+ " {\n",
+ " warning.style.display = \"block\";\n",
+ " }\n",
+ " if_use_prestored = document.getElementById(\"if_use_prestored\").checked;\n",
+ " var command = \" if_use_prestored_models = '\" + if_use_prestored.toString() + \"';\";\n",
+ " var kernel = IPython.notebook.kernel;\n",
+ " kernel.execute(command);\n",
+ " }\n",
+ "\n",
+ " function get_repr_regr_combination()\n",
+ " {\n",
+ " ngram_N = document.getElementById(\"input_ngram_N\").value;\n",
+ " n_threads = document.getElementById(\"input_n_threads\").value;\n",
+ " soap_cutoff = document.getElementById(\"input_soap_cutoff\").value;\n",
+ " soap_lmax = document.getElementById(\"input_soap_lmax\").value;\n",
+ " soap_nmax = document.getElementById(\"input_soap_nmax\").value;\n",
+ " \n",
+ "\n",
+ " \n",
+ " var command = \"ngram_N = int(\" + ngram_N + \"); n_threads = int(\" + n_threads + \");\" \n",
+ " + \" soap_cutoff = int(\" + soap_cutoff + \");\" \n",
+ " + \" soap_lmax = int(\" + soap_lmax + \");\" \n",
+ " + \" soap_nmax = int(\" + soap_nmax + \");\" ;\n",
+ " console.log(\"Executing Command: \" + command); \n",
+ " var kernel = IPython.notebook.kernel;\n",
+ " kernel.execute(command);\n",
+ " \n",
+ " \n",
+ " representation_value = document.getElementById(\"representation\").value;\n",
+ " regression_value = document.getElementById(\"regression\").value;\n",
+ " if_learning_curve = document.getElementById(\"if_learning_curve\").checked;\n",
+ " if_use_prestored = document.getElementById(\"if_use_prestored\").checked;\n",
+ " N_learning_curve = document.getElementById(\"N_learning_curve\").value;\n",
+ " \n",
+ " var command = \" representation = '\" + representation_value + \n",
+ " \"'; regression = '\" + regression_value + \"';\" + \" if_learningcurve = '\" + if_learning_curve.toString() + \"';\" \n",
+ " + \" if_use_prestored_models = '\" + if_use_prestored.toString() + \"';\"\n",
+ " + \" N_learning_curve = int(\" + N_learning_curve + \");\";\n",
+ " \n",
+ " console.log(\"Executing Command: \" + command);\n",
+ " \n",
+ " var kernel = IPython.notebook.kernel;\n",
+ " kernel.execute(command);\n",
+ "\n",
+ " Jupyter.notebook.execute_cells(window.findCellIndicesByTag('ngram'));//# N-gram descriptor\n",
+ " Jupyter.notebook.execute_cells(window.findCellIndicesByTag('soap'));//# SOAP descriptors\n",
+ " Jupyter.notebook.execute_cells(window.findCellIndicesByTag('krr-ngram'));//# KRR\n",
+ " Jupyter.notebook.execute_cells(window.findCellIndicesByTag('init')); \n",
+ " Jupyter.notebook.execute_cells(window.findCellIndicesByTag('init_read_data')); \n",
+ " Jupyter.notebook.execute_cells(window.findCellIndicesByTag('plot'));\n",
+ " Jupyter.notebook.execute_cells(window.findCellIndicesByTag('main'));//# Read and predict \n",
+ " \n",
+ " }\n",
+ " \n",
+ " \n",
+ "\n",
+ "</script>\n",
+ "\n"
+ ],
+ "text/plain": [
+ "<IPython.core.display.HTML object>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
"%%HTML\n",
"<br><br><br>\n",
@@ -888,18 +1554,31 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 5,
"metadata": {
"ExecuteTime": {
- "end_time": "2021-01-19T10:23:31.085760Z",
- "start_time": "2021-01-19T10:23:31.066074Z"
+ "end_time": "2021-11-11T17:10:52.773343Z",
+ "start_time": "2021-11-11T17:10:52.330752Z"
},
"init_cell": true,
"tags": [
"init"
]
},
- "outputs": [],
+ "outputs": [
+ {
+ "ename": "ModuleNotFoundError",
+ "evalue": "No module named 'networkx'",
+ "output_type": "error",
+ "traceback": [
+ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+ "\u001b[0;31mModuleNotFoundError\u001b[0m Traceback (most recent call last)",
+ "\u001b[0;32m<ipython-input-5-5decc5b9aed6>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m 11\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 12\u001b[0m \u001b[0;31m# n-gram\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 13\u001b[0;31m \u001b[0;32mimport\u001b[0m \u001b[0mkaggle_competition\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcrystal_graph\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mcg\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 14\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 15\u001b[0m \u001b[0;31m# SOAP\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;32m~/AI-toolkit/analytics/tutorials/analytics-kaggle-competition/kaggle_competition/__init__.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m 16\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mnumpy\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 17\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mpandas\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 18\u001b[0;31m \u001b[0;32mimport\u001b[0m \u001b[0mnetworkx\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mnx\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 19\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mmatplotlib\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpyplot\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 20\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mwarnings\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+ "\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'networkx'"
+ ]
+ }
+ ],
"source": [
"import os\n",
"import numpy as np\n",
@@ -2598,7 +3277,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.7.9"
+ "version": "3.7.3"
},
"toc": {
"base_numbering": 1,