diff --git a/beaker-notebooks/FF-fit.bkr b/beaker-notebooks/FF-fit.bkr
index c15e8616d962ed63cd254d0beca637df8cd7f622..f009b821f94fcb24d3c49bb25d1dd6cc514a40db 100644
--- a/beaker-notebooks/FF-fit.bkr
+++ b/beaker-notebooks/FF-fit.bkr
@@ -83,7 +83,7 @@
             "evaluator": "HTML",
             "input": {
                 "body": [
-                    "<a target=\"_blank\" href=\"http://forum.analytics-toolkit.nomad-coe.eu/\" class=\"btn btn-primary\"> Send your feedback to the analytics-toolkit forum</a><h2> Your comments are invaluable in helping us provide a user friendly experience for all! </h2>"
+                    "<a target=\"_blank\" href=\"https://nomad-forum.rz-berlin.mpg.de//\" class=\"btn btn-primary\"> Send your feedback to the analytics-toolkit forum</a><h2> Your comments are invaluable in helping us provide a user friendly experience for all! </h2>"
                 ],
                 "hidden": true
             },
@@ -94,7 +94,7 @@
                 "result": {
                     "type": "BeakerDisplay",
                     "innertype": "Html",
-                    "object": "<script>\nvar beaker = bkHelper.getBeakerObject().beakerObj;\n</script>\n<a target=\"_blank\" href=\"http://forum.analytics-toolkit.nomad-coe.eu/\" class=\"btn btn-primary\"> Send your feedback to the analytics-toolkit forum</a><h2> Your comments are invaluable in helping us provide a user friendly experience for all! </h2>"
+                    "object": "<script>\nvar beaker = bkHelper.getBeakerObject().beakerObj;\n</script>\n<a target=\"_blank\" href=\"https://nomad-forum.rz-berlin.mpg.de//\" class=\"btn btn-primary\"> Send your feedback to the analytics-toolkit forum</a><h2> Your comments are invaluable in helping us provide a user friendly experience for all! </h2>"
                 },
                 "elapsedTime": 0
             },
diff --git a/beaker-notebooks/LASSO_L0.bkr b/beaker-notebooks/LASSO_L0.bkr
index c628d6b85607c226d179ddec7f6c2cf313ba200b..37a06243e8bfa07e4810e78a85c1e6a3dcd77cd7 100644
--- a/beaker-notebooks/LASSO_L0.bkr
+++ b/beaker-notebooks/LASSO_L0.bkr
@@ -231,7 +231,7 @@
                     "</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;",
                     "",
                     "",
-                    "<a target=\"_blank\" href=\"http://forum.analytics-toolkit.nomad-coe.eu/\" class=\"btn btn-primary\"> Tell us what you think</a>"
+                    "<a target=\"_blank\" href=\"https://nomad-forum.rz-berlin.mpg.de//\" class=\"btn btn-primary\"> Tell us what you think</a>"
                 ],
                 "hidden": true
             },
@@ -240,7 +240,7 @@
                 "result": {
                     "type": "BeakerDisplay",
                     "innertype": "Html",
-                    "object": "<script>\nvar beaker = bkHelper.getBeakerObject().beakerObj;\n</script>\n<script>\nvar run_lasso = function() {\n  $(\"#lasso_result_button\").removeClass(\"active\").addClass(\"disabled\");\n  getFeatures();\n  getOperators();\n  getMaxDim();\n  getUnits();\n  beaker.evaluate(\"calc_cell\"); // evaluate cells with tag \"calc_cell\"\n // view_result()\n};\nvar reset_lasso = function(){\n  beaker.evaluate(\"lasso-settings-cell\");\n  var e = document.getElementById('lasso-hidden-settings-div');\n  var b = document.getElementById('lasso-hidden-settings-button');\n  e.style.display = 'block';\n  b.style.display = 'inline';\n};\nvar getFeatures = function() {\n    beaker.selected_feature_list = [];\n    $('#lasso_features_select input:checkbox').each(function () {\n        if(this.checked )\n          beaker.selected_feature_list.push(this.value);\n    });\n};\nvar getOperators = function() {\n    beaker.allowed_operations = [];\n    $('#lasso_operators_select input:checkbox').each(function () {\n        if(this.checked )\n          beaker.allowed_operations.push(this.value);\n    });\n};  \nvar getMaxDim = function() {\n   beaker.max_dim = $(\"#lasso_max_dim_selector\").val();\n};\n  \nvar getUnits = function() {\n   beaker.units = $(\"#units_select\").val();\n};\nvar toggle_settings = function(){\n  var e = document.getElementById('lasso-hidden-settings-div');\n  var b = document.getElementById('lasso-hidden-settings-button');\n  if(e.style.display == 'block'){\n    e.style.display = 'none';\n    b.style.display = 'none';\n  }\n  else{\n    e.style.display = 'block';\n    b.style.display = 'inline';\n  }\n};\nbeaker.view_result = function(result_link) {\n//   beaker.evaluate(\"lasso_viewer_result\").then(function(x) {\n    $(\"#lasso_result_button\").attr(\"href\", result_link);\n//   }); \n  $(\"#lasso_result_button\").removeClass(\"disabled\").addClass(\"active\");\n}\n</script>\n<style type=\"text/css\">\n .lasso_instructions{\n    font-size: 15px;\n  } \n</style>\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" data-toggle=\"modal\" data-target=\"#lasso-motivation-modal\">\n Background\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<!-- Modal -->\n<div style=\"display: none;\" class=\"modal fade\" id=\"lasso-motivation-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"lasso-motivation-modal-label\">\n  <div class=\"modal-dialog modal-lg\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"lasso-motivation-modal-label\">Background</h4>\n      </div>\n      <div class=\"modal-body lasso_instructions\">\n        <p> In this tutorial we present a tool for predicting the crystal structure of octet binary compounds, by using a set of descriptive parameters (a descriptor) based on free-atom data of the atomic species constituting the binary material.</p>\n\n        <p>In this example, we address only Rocksalt (RS) and Zincblende (ZB) crystal structures, that are the most common for the material class of octet binaries. Specifically, the tool predicts the difference in total energy between RS and ZB equilibrated structures (i.e., each structure is relaxed to its local minimum).</p>\n\n        <p>The prediction of RS vs ZB structure from a simple descriptor has a notable history in materials science [1-7], where descriptors were designed by chemically/physically-inspired intuition. The tool presented here allows for the machine-learning-aided automatic discovery of a descriptor and a model for the prediction of the difference in energy between RS and ZB for 82 octet binary materials.</p>\n\n        <p>The tool is based on Compressed-sensing (LASSO performed on a tailor-made feature space, followed by L0-regularized minimization, click <a href=\"https://gitlab.rzg.mpg.de/nomad-lab/public-wiki/wikis/analytics/LASSO_L0\" target=\"_blank\">here</a> for more info on the LASSO+L0 method), as introduced in:  </p>\n\n        <p> \"Big Data of Materials Science: Critical Role of the Descriptor\". L. M. Ghiringhelli, J. Vybiral, S. V. Levchenko, C. Draxl, and M. Scheffler Phys. Rev. Lett. 114, 105503 (2015) <a href=\"http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.114.105503\" target=\"_blank\"> (Click here for the free access pdf) </a></p>\n\n        <p> By running the tutorial with the default setting, the results of the <a href=\"http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.114.105503\" target=\"_blank\">PRL 2015</a> can be recovered. In particular, by clicking on “View interactive 2D plot”, an interactive structure-map (a chart where different structures are located in different regions of a low-dimensional representation, here two dimensional) will be opened in a new tab, similar to the following (an extended version of Fig. 2 in <a href=\"http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.114.105503\" target=\"_blank\">PRL 2015</a>): </p>\n        \n        <img style=\"width:67%;height:67%\" src=\"https://gitlab.mpcdf.mpg.de/nomad-lab/public-wiki/uploads/eb33f1415db1b6d489cbbc3e6a899942/2016-08-02_ZB_RS3-2.png\">\n        <br>\n        <br>\n        <p> In this map the octet binaries are located via the descriptor found by our LASSO+L0 approach. The descriptor is based purely on free-atom data, namely radii of the <i>s</i> and <i>p</i> valence orbitals (rs and rp) of the atomic species and their Ionizaiton Potential and Electron Affinity (IP and EA). Materials in the red (blue) region crystallize preferably in the zincblende (rocksalt) structure. The distance to the green line is proportional to the difference in energy between the two structures. In the interactive plot accessible at the end of the learning performed by the present tool, one can obtain information on the materials by hovering and clicking on the data points. </p>\n        <p>References:</p>\n        <ol>\n          <li>J. A. van Vechten, Phys. Rev. 182, 891 (1969).</li>\n          <li>J. C. Phillips, Rev. Mod. Phys. 42, 317 (1970).</li>\n          <li>J. St. John and A.N. Bloch, Phys. Rev. Lett. 33, 1095 (1974).</li>\n          <li>J. R. Chelikowsky and J. C. Phillips, Phys. Rev. B 17, 2453 (1978).</li>\n          <li>A. Zunger, Phys. Rev. B 22, 5839 (1980).</li>\n          <li>D. G. Pettifor, Solid State Commun. 51, 31 (1984).</li>\n          <li>Y. Saad, D. Gao, T. Ngo, S. Bobbitt, J. R. Chelikowsky, and W. Andreoni, Phys. Rev. B 85, 104104 (2012).</li>\n        </ol>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n<!--         <button type=\"button\" class=\"btn btn-primary\">Save changes</button> -->\n      </div>\n    </div>\n  </div>\n</div>\n\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" data-toggle=\"modal\" data-target=\"#lasso-instructions-modal\">\n Instructions\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<!-- Modal -->\n<div class=\"modal fade\" id=\"lasso-instructions-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"lasso-instructions-modal-label\">\n  <div class=\"modal-dialog\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"lasso-instructions-modal-label\">Instructions</h4>\n      </div>\n      <div class=\"modal-body lasso_instructions\">\n<p>In this example, you can run a compressed-sensing based algorithm for finding the optimal descriptor (and model) that predicts the difference in energy between crystal structures (here, rocksalt vs. zincblende). </p>\n\n<p>The descriptor is selected out of a large number of candidates constructed as functions of basic input features, the primary features. </p>\n\n<p>You can select the primary features as well as which kind of unary and binary operations are allowed from the checklist below. You can also select the maximum dimensionality of the descriptor. </p>\n<p>        After the wished features have been selected, click <b>RUN</b> to perform the calculations (loading the values of the primary features, creation of the feature space, and optimization via LASSO+L0). </p>\n\nDuring the run, a brief summary is printed out below the <b>RUN</b> button. At the end of the run: \n  <ul>\n  <li> the solution (machine-learned descriptor, model, and its performance in terms of training error) is printed out underneath starting from the one-dimensional solution to the selected maximum dimensionality and</li>\n<li> the “View interactive 2D scatter plot” button unlocks; by clicking, the scatter plot with the two-dimensional descriptor appears in a separate tab. In case a dimensionality higher than 2 was selected for the descriptor, the plot displays the first two dimensions.</li>\n</ul>\n<p>Note: the plot stays active also after another run is performed, so that the output of several sets of input parameters can be compared in the viewer tabs.</p>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n<!--         <button type=\"button\" class=\"btn btn-primary\">Save changes</button> -->\n      </div>\n    </div>\n  </div>\n</div>\n\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" onclick=\"toggle_settings()\">\n Settings\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n\n<a target=\"_blank\" href=\"http://forum.analytics-toolkit.nomad-coe.eu/\" class=\"btn btn-primary\"> Tell us what you think</a>"
+                    "object": "<script>\nvar beaker = bkHelper.getBeakerObject().beakerObj;\n</script>\n<script>\nvar run_lasso = function() {\n  $(\"#lasso_result_button\").removeClass(\"active\").addClass(\"disabled\");\n  getFeatures();\n  getOperators();\n  getMaxDim();\n  getUnits();\n  beaker.evaluate(\"calc_cell\"); // evaluate cells with tag \"calc_cell\"\n // view_result()\n};\nvar reset_lasso = function(){\n  beaker.evaluate(\"lasso-settings-cell\");\n  var e = document.getElementById('lasso-hidden-settings-div');\n  var b = document.getElementById('lasso-hidden-settings-button');\n  e.style.display = 'block';\n  b.style.display = 'inline';\n};\nvar getFeatures = function() {\n    beaker.selected_feature_list = [];\n    $('#lasso_features_select input:checkbox').each(function () {\n        if(this.checked )\n          beaker.selected_feature_list.push(this.value);\n    });\n};\nvar getOperators = function() {\n    beaker.allowed_operations = [];\n    $('#lasso_operators_select input:checkbox').each(function () {\n        if(this.checked )\n          beaker.allowed_operations.push(this.value);\n    });\n};  \nvar getMaxDim = function() {\n   beaker.max_dim = $(\"#lasso_max_dim_selector\").val();\n};\n  \nvar getUnits = function() {\n   beaker.units = $(\"#units_select\").val();\n};\nvar toggle_settings = function(){\n  var e = document.getElementById('lasso-hidden-settings-div');\n  var b = document.getElementById('lasso-hidden-settings-button');\n  if(e.style.display == 'block'){\n    e.style.display = 'none';\n    b.style.display = 'none';\n  }\n  else{\n    e.style.display = 'block';\n    b.style.display = 'inline';\n  }\n};\nbeaker.view_result = function(result_link) {\n//   beaker.evaluate(\"lasso_viewer_result\").then(function(x) {\n    $(\"#lasso_result_button\").attr(\"href\", result_link);\n//   }); \n  $(\"#lasso_result_button\").removeClass(\"disabled\").addClass(\"active\");\n}\n</script>\n<style type=\"text/css\">\n .lasso_instructions{\n    font-size: 15px;\n  } \n</style>\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" data-toggle=\"modal\" data-target=\"#lasso-motivation-modal\">\n Background\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<!-- Modal -->\n<div style=\"display: none;\" class=\"modal fade\" id=\"lasso-motivation-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"lasso-motivation-modal-label\">\n  <div class=\"modal-dialog modal-lg\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"lasso-motivation-modal-label\">Background</h4>\n      </div>\n      <div class=\"modal-body lasso_instructions\">\n        <p> In this tutorial we present a tool for predicting the crystal structure of octet binary compounds, by using a set of descriptive parameters (a descriptor) based on free-atom data of the atomic species constituting the binary material.</p>\n\n        <p>In this example, we address only Rocksalt (RS) and Zincblende (ZB) crystal structures, that are the most common for the material class of octet binaries. Specifically, the tool predicts the difference in total energy between RS and ZB equilibrated structures (i.e., each structure is relaxed to its local minimum).</p>\n\n        <p>The prediction of RS vs ZB structure from a simple descriptor has a notable history in materials science [1-7], where descriptors were designed by chemically/physically-inspired intuition. The tool presented here allows for the machine-learning-aided automatic discovery of a descriptor and a model for the prediction of the difference in energy between RS and ZB for 82 octet binary materials.</p>\n\n        <p>The tool is based on Compressed-sensing (LASSO performed on a tailor-made feature space, followed by L0-regularized minimization, click <a href=\"https://gitlab.rzg.mpg.de/nomad-lab/public-wiki/wikis/analytics/LASSO_L0\" target=\"_blank\">here</a> for more info on the LASSO+L0 method), as introduced in:  </p>\n\n        <p> \"Big Data of Materials Science: Critical Role of the Descriptor\". L. M. Ghiringhelli, J. Vybiral, S. V. Levchenko, C. Draxl, and M. Scheffler Phys. Rev. Lett. 114, 105503 (2015) <a href=\"http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.114.105503\" target=\"_blank\"> (Click here for the free access pdf) </a></p>\n\n        <p> By running the tutorial with the default setting, the results of the <a href=\"http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.114.105503\" target=\"_blank\">PRL 2015</a> can be recovered. In particular, by clicking on “View interactive 2D plot”, an interactive structure-map (a chart where different structures are located in different regions of a low-dimensional representation, here two dimensional) will be opened in a new tab, similar to the following (an extended version of Fig. 2 in <a href=\"http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.114.105503\" target=\"_blank\">PRL 2015</a>): </p>\n        \n        <img style=\"width:67%;height:67%\" src=\"https://gitlab.mpcdf.mpg.de/nomad-lab/public-wiki/uploads/eb33f1415db1b6d489cbbc3e6a899942/2016-08-02_ZB_RS3-2.png\">\n        <br>\n        <br>\n        <p> In this map the octet binaries are located via the descriptor found by our LASSO+L0 approach. The descriptor is based purely on free-atom data, namely radii of the <i>s</i> and <i>p</i> valence orbitals (rs and rp) of the atomic species and their Ionizaiton Potential and Electron Affinity (IP and EA). Materials in the red (blue) region crystallize preferably in the zincblende (rocksalt) structure. The distance to the green line is proportional to the difference in energy between the two structures. In the interactive plot accessible at the end of the learning performed by the present tool, one can obtain information on the materials by hovering and clicking on the data points. </p>\n        <p>References:</p>\n        <ol>\n          <li>J. A. van Vechten, Phys. Rev. 182, 891 (1969).</li>\n          <li>J. C. Phillips, Rev. Mod. Phys. 42, 317 (1970).</li>\n          <li>J. St. John and A.N. Bloch, Phys. Rev. Lett. 33, 1095 (1974).</li>\n          <li>J. R. Chelikowsky and J. C. Phillips, Phys. Rev. B 17, 2453 (1978).</li>\n          <li>A. Zunger, Phys. Rev. B 22, 5839 (1980).</li>\n          <li>D. G. Pettifor, Solid State Commun. 51, 31 (1984).</li>\n          <li>Y. Saad, D. Gao, T. Ngo, S. Bobbitt, J. R. Chelikowsky, and W. Andreoni, Phys. Rev. B 85, 104104 (2012).</li>\n        </ol>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n<!--         <button type=\"button\" class=\"btn btn-primary\">Save changes</button> -->\n      </div>\n    </div>\n  </div>\n</div>\n\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" data-toggle=\"modal\" data-target=\"#lasso-instructions-modal\">\n Instructions\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<!-- Modal -->\n<div class=\"modal fade\" id=\"lasso-instructions-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"lasso-instructions-modal-label\">\n  <div class=\"modal-dialog\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"lasso-instructions-modal-label\">Instructions</h4>\n      </div>\n      <div class=\"modal-body lasso_instructions\">\n<p>In this example, you can run a compressed-sensing based algorithm for finding the optimal descriptor (and model) that predicts the difference in energy between crystal structures (here, rocksalt vs. zincblende). </p>\n\n<p>The descriptor is selected out of a large number of candidates constructed as functions of basic input features, the primary features. </p>\n\n<p>You can select the primary features as well as which kind of unary and binary operations are allowed from the checklist below. You can also select the maximum dimensionality of the descriptor. </p>\n<p>        After the wished features have been selected, click <b>RUN</b> to perform the calculations (loading the values of the primary features, creation of the feature space, and optimization via LASSO+L0). </p>\n\nDuring the run, a brief summary is printed out below the <b>RUN</b> button. At the end of the run: \n  <ul>\n  <li> the solution (machine-learned descriptor, model, and its performance in terms of training error) is printed out underneath starting from the one-dimensional solution to the selected maximum dimensionality and</li>\n<li> the “View interactive 2D scatter plot” button unlocks; by clicking, the scatter plot with the two-dimensional descriptor appears in a separate tab. In case a dimensionality higher than 2 was selected for the descriptor, the plot displays the first two dimensions.</li>\n</ul>\n<p>Note: the plot stays active also after another run is performed, so that the output of several sets of input parameters can be compared in the viewer tabs.</p>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n<!--         <button type=\"button\" class=\"btn btn-primary\">Save changes</button> -->\n      </div>\n    </div>\n  </div>\n</div>\n\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" onclick=\"toggle_settings()\">\n Settings\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n\n<a target=\"_blank\" href=\"https://nomad-forum.rz-berlin.mpg.de//\" class=\"btn btn-primary\"> Tell us what you think</a>"
                 },
                 "selectedType": "BeakerDisplay",
                 "elapsedTime": 0,
diff --git a/beaker-notebooks/brprototype3.bkr b/beaker-notebooks/brprototype3.bkr
index b25a8111738d2354ce7ffb1d63045d39a160d0dd..dc51dbc721140b7145416f97bc343961c212517b 100755
--- a/beaker-notebooks/brprototype3.bkr
+++ b/beaker-notebooks/brprototype3.bkr
@@ -155,7 +155,7 @@
                     "  <button type=\"button\" class=\"btn btn-primary\" style=\"margin-top: 2ex;\" onclick=\"trainAndPredict();\">Build model!</button>&nbsp;&nbsp;",
                     "  <button type=\"button\" class=\"btn btn-secondary\" style=\"margin-top: 2ex;\" data-toggle=\"modal\" data-target=\"#qmmldemo-explanation-modal\">Explanation</button>&nbsp;&nbsp;",
                     "  <button type=\"button\" class=\"btn btn-secondary\" style=\"margin-top: 2ex;\" onclick='toggle_settings()'>Settings</button>&nbsp;&nbsp;",
-                    "  <a target=\"_blank\" href=\"http://forum.analytics-toolkit.nomad-coe.eu/\"><button type=\"button\" class=\"btn btn-primary\" style=\"margin-top: 2ex;\">Tell us what you think</button></a>    ",
+                    "  <a target=\"_blank\" href=\"https://nomad-forum.rz-berlin.mpg.de//\"><button type=\"button\" class=\"btn btn-primary\" style=\"margin-top: 2ex;\">Tell us what you think</button></a>    ",
                     "</div>",
                     "",
                     "<script>",
@@ -440,7 +440,7 @@
                 "result": {
                     "type": "BeakerDisplay",
                     "innertype": "Html",
-                    "object": "<script>\nvar beaker = bkHelper.getBeakerObject().beakerObj;\n</script>\n<!-- Header -->\n\n<div style=\"font-size: 250%; font-weight: bold;\">Accurate predictions of molecular properties</div>\n\n<!-- <div style=\"font-size: 150%; margin-bottom: 1em;\">A <a href=\"https://nomad-coe.eu/\">NOMAD</a> demonstration by <a href=\"http://mrupp.info/\">Matthias Rupp</a>, 2016.</div> -->\n<div style=\"font-size: 140%; margin-bottom: 1em;\"><a href=\"http://mrupp.info/\">Matthias Rupp</a> <span style=\"font-size: 65%;\">[version 2017-01-25]</span></div>\n\n<div style=\"max-width: 600px;\">\n  This tutorial demonstrates prediction of atomization energies and other properties of small organic molecules at the level of hybrid density functional theory and others, based on nuclear charges and atomic positions only.\n  It is based on\n</div>\n<div style=\"padding: 1ex; margin-top: 1ex; margin-bottom: 1ex; border-style: dotted; border-width: 1pt; border-color: blue; border-radius: 3px; max-width: 600px;\">\n  <p>M. Rupp, A. Tkatchenko, K.-R. Müller, O.A. von Lilienfeld: <span style=\"font-style: italic;\">Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning</span>, Phys. Rev. Lett. 108: 058301 (2012). <a href=\"https://doi.org/10.1103/PhysRevLett.108.058301\">[DOI]</a></p>\n  <!--<p>G. Montavon, M. Rupp, V. Gobre, A. Vazquez-Mayagoitia, K. Hansen, A. Tkatchenko, K.-R. Müller, O.A. von Lilienfeld: <span style=\"font-style: italic;\">Machine learning of molecular electronic properties in chemical compound space</span>, New J. Phys. 15(9): 095003 (2013). <a href=\"http://dx.doi.org/10.1088/1367-2630/15/9/095003\">[DOI]</a></p>-->\n  M. Rupp: <span style=\"font-style: italic;\">Machine Learning for Quantum Mechanics in a Nutshell</span>, Int. J. Quant. Chem. 115(16): 1058-1073 (2015). <a href=\"http://dx.doi.org/10.1002/qua.24954\">[DOI]</a>\n</div>\n<div style=\"max-width: 600px;\">\n  Click on \"Build model!\" below to build a model for atomization energy; click \"Explanation\" for an exposition of the approach; or modify \"Settings\" to produce your own results.\n</div>\n\n<div style=\"padding-top: 2ex; max-width: 600px;\">\n  <span style=\"font-weight: bold;\">Idea: </span> \n  <!--Machine learning can be used to rapidly and accurately predict outcomes of electronic structure calculations.-->\n  To predict ab initio properties across molecules, the problem of solving the molecular Schrödinger equation is mapped onto a nonlinear statistical regression problem.\n  Each electronic structure calculation thus becomes a training point for the regression.\n  The trick is how to represent molecules in an abstract space that supports interpolation.\n</div>\n\n<div style=\"top-margin: 3ex;\">\n  <button type=\"button\" class=\"btn btn-primary\" style=\"margin-top: 2ex;\" onclick=\"trainAndPredict();\">Build model!</button>&nbsp;&nbsp;\n  <button type=\"button\" class=\"btn btn-secondary\" style=\"margin-top: 2ex;\" data-toggle=\"modal\" data-target=\"#qmmldemo-explanation-modal\">Explanation</button>&nbsp;&nbsp;\n  <button type=\"button\" class=\"btn btn-secondary\" style=\"margin-top: 2ex;\" onclick=\"toggle_settings()\">Settings</button>&nbsp;&nbsp;\n  <a target=\"_blank\" href=\"http://forum.analytics-toolkit.nomad-coe.eu/\"><button type=\"button\" class=\"btn btn-primary\" style=\"margin-top: 2ex;\">Tell us what you think</button></a>    \n</div>\n\n<script>\n  // Shows/hides settings block\n  var toggle_settings = function() {\n    var e = document.getElementById('qmmldemo-hidden-settings-div');\n    if(e.style.display == 'block') { e.style.display = 'none'; } else { e.style.display = 'block'; };\n  }\n  \n  // Adds an option to a dropdown menu\n  function addDropdownChoice(dropdown, value, content) {\n    var el = document.createElement('option');\n    el.value = value;\n    el.innerHTML = content\n    dropdown.appendChild(el);\n  }\n  \n  // Adds an option to a radiobutton selector\n  function addRadiobuttonChoice(radiobutton, value, name, content, checked) {\n    var el = document.createElement('input');\n    el.type = \"radio\";\n    el.value = value;\n    el.name = name;\n    if(checked) { el.checked = \"checked\" }\n\n    var label = document.createElement('label');\n    label.innerHTML = content; label.innerHTML += ' ';\n    label.appendChild(el);\n\n    radiobutton.appendChild(label);\n  }\n  \n  function qmmlUpdateForm() {\n    var ds = document.getElementById(\"qmmldemo_dataset\").value;\n    \n    var dsd = document.getElementById(\"qmmldemo_dataset_description\");\n    var p   = document.getElementById(\"qmmldemo_property\"); p.innerHTML = '';\n    var td  = document.getElementById(\"qmmldemo_traindata\"); td.innerHTML = '';\n\n    switch(ds) {\n      case \"1\":  // PRL 2012\n        dsd.innerHTML = '   M.&nbsp;Rupp <i>et&nbsp;al</i>, Phys. Rev. Lett. 108(5): 058301, 2012  <a href=\"http://dx.doi.org/10.1103/PhysRevLett.108.058301\">DOI</a>';\n\n        addDropdownChoice(p, \"1\", \"atomization energy (DFT/PBE0)\");\n\n        addRadiobuttonChoice(td, \"1000\", \"qmmldemo_traindata_radio\", \"1k\", true );\n        addRadiobuttonChoice(td, \"3000\", \"qmmldemo_traindata_radio\", \"2k\", false);\n        addRadiobuttonChoice(td, \"5000\", \"qmmldemo_traindata_radio\", \"5k\", false);\n        \n        break;\n        \n      case \"2\":  // New J Phys 2013\n        dsd.innerHTML = '   G.&nbsp;Montavon <i>et&nbsp;al</i>, New J. Phys. 15(9): 095003, 2013  <a href=\"http://dx.doi.org/10.1088/1367-2630/15/9/095003\">DOI</a>';\n        \n        addDropdownChoice(p, \"1\", \"atomization energy (DFT/PBE0)\");\n        addDropdownChoice(p, \"2\", \"static polarizability (DFT/PBE0)\");\n        addDropdownChoice(p, \"3\", \"HOMO (DFT/PBE0)\");\n        addDropdownChoice(p, \"4\", \"HOMO (GW)\");\n        addDropdownChoice(p, \"5\", \"LUMO (DFT/PBE0)\");\n        addDropdownChoice(p, \"6\", \"LUMO (GW)\");\n\n        addRadiobuttonChoice(td, \"1000\", \"qmmldemo_traindata_radio\", \"1k\", true );\n        addRadiobuttonChoice(td, \"3000\", \"qmmldemo_traindata_radio\", \"2k\", false);\n        addRadiobuttonChoice(td, \"5000\", \"qmmldemo_traindata_radio\", \"5k\", false);\n        \n        break;\n\n      case \"3\":  //  Sci Data 2014\n        dsd.innerHTML = '   R.&nbsp;Ramakrishnan <i>et&nbsp;al</i>, Sci. Data 1: 140022, 2014  <a href=\"http://dx.doi.org/10.1038/sdata.2014.22\">DOI</a>';\n\n        addDropdownChoice(p, \"1\", \"atomization energy\");\n        \n        addRadiobuttonChoice(td,  \"1000\", \"qmmldemo_traindata_radio\",  \"1k\", true );\n        addRadiobuttonChoice(td,  \"3000\", \"qmmldemo_traindata_radio\",  \"2k\", false);\n        addRadiobuttonChoice(td,  \"5000\", \"qmmldemo_traindata_radio\",  \"5k\", false);\n        addRadiobuttonChoice(td, \"10000\", \"qmmldemo_traindata_radio\", \"10k\", false);\n\n        break;\n        \n      case \"4\":  // Sci Data 2014\n        dsd.innerHTML = '   R.&nbsp;Ramakrishnan <i>et&nbsp;al</i>, Sci. Data 1: 140022, 2014  <a href=\"http://dx.doi.org/10.1038/sdata.2014.22\">DOI</a>';\n\n        addDropdownChoice(p, \"1\", \"zero point vibrational energy (G4MP2)\");\n        addDropdownChoice(p, \"2\", \"internal energy @ 0 K (G4MP2)\");\n        addDropdownChoice(p, \"3\", \"internal energy @ 298.15 K (G4MP2)\");\n        addDropdownChoice(p, \"4\", \"enthalpy @ 298.15 K (G4MP2)\");\n        addDropdownChoice(p, \"5\", \"free energy @ 298.15 K (G4MP2)\");\n        addDropdownChoice(p, \"6\", \"heat capacity @ 298.15 K (DFT/B3LYP)\");\n\n        addRadiobuttonChoice(td, \"1000\", \"qmmldemo_traindata_radio\", \"1k\", true );\n        addRadiobuttonChoice(td, \"3000\", \"qmmldemo_traindata_radio\", \"2k\", false);\n        addRadiobuttonChoice(td, \"5000\", \"qmmldemo_traindata_radio\", \"5k\", false);\n\n        break;\n    }\n  }\n\n  function qmmlUpdateProperty() {\n    var ds  = document.getElementById(\"qmmldemo_dataset\").value;\n    var p   = document.getElementById(\"qmmldemo_property\").value;\n    var pd  = document.getElementById(\"qmmldemo_property_description\");\n\n    switch(ds) {\n      case \"1\": switch(p) {\n        case \"1\": pd.innerHTML = '   atomization energy, DFT/PBE0'; break;\n      }; break;\n      case \"2\": switch(p) {\n        case \"1\": pd.innerHTML = '   atomization energy, DFT/PBE0'; break;\n        case \"2\": pd.innerHTML = '   averaged static polarizability, DFT/PBE0'; break;\n        case \"3\": pd.innerHTML = '   highest occupied molecular orbital, DFT/PBE0'; break;\n        case \"4\": pd.innerHTML = '   highest occupied molecular orbital, GW'; break; \n        case \"5\": pd.innerHTML = '   lowest unoccupied molecular orbital, DFT/PBE0'; break; \n        case \"6\": pd.innerHTML = '   lowest unoccupied molecular orbital, GW'; break; \n      }; break;\n      case \"3\": switch(p) {\n        case \"1\": pd.innerHTML = '   atomization energy, DFT/B3LYP/6-31G(2df,p)'; break;\n      }; break;\n      case \"4\": switch(p) {\n        case \"1\": pd.innerHTML = '   zero point vibrational energy, G4MP2'; break;\n        case \"2\": pd.innerHTML = '   internal energy at 0 K, G4MP2'; break;\n        case \"3\": pd.innerHTML = '   internal energy at 298.15 K, G4MP2'; break;\n        case \"4\": pd.innerHTML = '   enthalpy at 298.15 K, G4MP2'; break;\n        case \"5\": pd.innerHTML = '   free energy at 298.15 K, G4MP2'; break;\n        case \"6\": pd.innerHTML = '   heat capacity at 298.15 K, DFT/B3LYP'; break;\n      }; break;\n    }\n  }\n\n  function qmmlUpdateRepresentation() {\n    var r   = document.getElementById(\"qmmldemo_representation\").value;\n    var rd  = document.getElementById(\"qmmldemo_representation_description\");\n    switch(r) {\n      case \"1\": rd.innerHTML = '   M.&nbsp;Rupp <i>et&nbsp;al</i>, Phys. Rev. Lett. 108(5): 058301, 2012  <a href=\"http://dx.doi.org/10.1103/PhysRevLett.108.058301\">DOI</a>'; break;\n      case \"2\": rd.innerHTML = '   M.&nbsp;Rupp <i>et&nbsp;al</i>, Phys. Rev. Lett. 108(5): 058301, 2012  <a href=\"http://dx.doi.org/10.1103/PhysRevLett.108.058301\">DOI</a>'; break;\n      case \"3\": rd.innerHTML = '   K.&nbsp;Hansen <i>et&nbsp;al</i>, J. Phys. Chem. Lett. 6: 2326, 2015  <a href=\"http://dx.doi.org/10.1021/acs.jpclett.5b00831\">DOI</a>'; break;\n    }\n  }\n  \n  function qmmlUpdateKernel() {\n    var k  = document.getElementById(\"qmmldemo_kernel\").value;\n    var kd = document.getElementById(\"qmmldemo_kernel_description\");\n    switch(k) {\n      case \"1\": kd.innerHTML = \"   linear basis functions, no free parameters\"; break;\n      case \"2\": kd.innerHTML = \"   Gaussian basis functions, length scale free parameter\"; break;\n      case \"3\": kd.innerHTML = \"   exponential basis functions, length scale free parameter\"; break;\n    }\n  }\n  \n  function trainAndPredict() {\n    beaker.ctrl_ds   = document.getElementById(\"qmmldemo_dataset\").value;\n    beaker.ctrl_pty  = document.getElementById(\"qmmldemo_property\").value;\n    beaker.ctrl_repr = document.getElementById(\"qmmldemo_representation\").value;\n    beaker.ctrl_k    = document.getElementById(\"qmmldemo_kernel\").value;\n    beaker.ctrl_n    = document.querySelector('input[name=\"qmmldemo_traindata_radio\"]:checked').value;\n    beaker.evaluate(\"trainAndPredictCell\");\n  }\n</script>\n\n<style type=\"text/css\">\n  \n  .qmmldemo_table th { font-weight: bold; padding-right: 2ex; }\n  .qmmldemo_table td input { margin-right: 1ex; }\n  \n</style>\n\n<!-- Controls area -->\n\n<div class=\"qmmldemo_control\" id=\"qmmldemo-hidden-settings-div\" style=\"display: none;\">\n  <p style=\"height: 2ex;\"></p>\n  <table class=\"qmmldemo_table\">\n    \n    <tbody><tr><th>Dataset:</th>\n      <td>\n        <select id=\"qmmldemo_dataset\" onchange=\"qmmlUpdateForm()\">\n          <option value=\"1\" selected=\"\">7k small organic molecules (GDB7-12)</option>\n          <option value=\"2\">7k small organic molecules (GDB7-13)</option>\n          <!-- <option value=\"3\">134k small organic molecules (GDB9-14)</option> -->\n          <option value=\"4\">6k C7H20O2 isomers (GDB9-14)</option>\n        </select>\n      </td>\n      <td id=\"qmmldemo_dataset_description\" style=\"white-space: pre;\">   M.&nbsp;Rupp <i>et&nbsp;al</i>, Phys. Rev. Lett. 108(5): 058301, 2012  <a href=\"http://dx.doi.org/10.1103/PhysRevLett.108.058301\">DOI</a></td>\n    </tr>\n    \n    <tr>\n      <th>Property:</th>\n      <td><select id=\"qmmldemo_property\" onchange=\"qmmlUpdateProperty()\"><option value=\"1\">atomization energy (DFT/PBE0)</option></select></td>\n      <td id=\"qmmldemo_property_description\" style=\"white-space: pre;\">   atomization energy, DFT/PBE0</td>\n    </tr>\n    \n    <tr>\n      <th>Representation:</th>\n      <td><select id=\"qmmldemo_representation\" onchange=\"qmmlUpdateRepresentation()\">\n        <option value=\"1\">sorted Coulomb matrix</option>\n        <option value=\"2\">diagonalized Coulomb matrix</option>\n        <!-- <option value=\"3\">bag of bonds</option> -->\n       </select></td>\n      <td id=\"qmmldemo_representation_description\" style=\"white-space: pre;\">   M.&nbsp;Rupp <i>et&nbsp;al</i>, Phys. Rev. Lett. 108(5): 058301, 2012  <a href=\"http://dx.doi.org/10.1103/PhysRevLett.108.058301\">DOI</a></td>\n    </tr>\n    \n    <tr>\n      <th>Kernel:</th>\n      <td><select id=\"qmmldemo_kernel\" onchange=\"qmmlUpdateKernel()\">\n        <option value=\"1\">linear kernel</option>\n        <option value=\"2\">Gaussian kernel</option>\n        <option value=\"3\">Laplacian kernel</option>         \n      </select></td>\n      <td id=\"qmmldemo_kernel_description\" style=\"white-space: pre;\">   linear basis functions, no free parameters</td>\n    </tr>\n    \n    <tr>\n      <th>Training data:</th>\n      <td><form id=\"qmmldemo_traindata\"><label>1k <input value=\"1000\" name=\"qmmldemo_traindata_radio\" type=\"radio\"></label><label>2k <input value=\"3000\" name=\"qmmldemo_traindata_radio\" type=\"radio\"></label><label>5k <input value=\"5000\" name=\"qmmldemo_traindata_radio\" type=\"radio\"></label></form></td>\n      <td style=\"white-space: pre;\">   number of training examples</td>\n    </tr>\n  </tbody></table>\n</div>\n\n<!-- Explanation -->\n<div class=\"modal fade\" id=\"qmmldemo-explanation-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"qmmldemo-explanation-modal-label\" style=\"display: none;\">\n  <div class=\"modal-dialog modal-lg\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"qmmldemo-explanation-modal-label\">Explanation</h4>\n      </div>\n      <div class=\"modal-body qmmldemo_explanation\">\n        <p>\n          In this tutorial we demonstrate fast and accurate prediction of ab initio properties of small organic molecules.\n        </p>\n        <p>\n          The idea is that computing the same property for many similar molecules entails a high degree of redundancy:\n          The same deterministic numeric procedure is carried out repeatedly for correlated inputs, yielding correlated outputs.\n          Consequently, machine learning can be used to interpolate between reference calculations, exploiting this redundancy.\n        </p>\n        <p>\n          In such models, the problem of repeatedly solving the electronic Schrödinger equation is mapped onto a non-linear regression problem.\n          The key challenge is to define a space (or basis set) in which each reference calculation is a point and that enables efficient interpolation.\n          If successful, the interpolation / prediction error decreases with the number of reference calculations (training data):\n        </p>\n        <img src=\"http://iopscience.iop.org/1367-2630/15/9/095003/downloadHRFigure/figure/nj459085f2\" alt=\"learning curve\" style=\"margin: 2ex; width: 60%; padding-left: 3ex;\">\n        <p>\n          Here, MAE = mean absolute error over a validation set; predictions are for E = atomization energy, α = polarizability.\n          The machine learning predictions themselves have negligible computational cost (on the order of milliseconds) compared to electronic structure calculations.\n        </p>\n        <p>\n          This tutorial aims at demonstrating interpolation across changes in chemistry.\n          Molecules are therefore assumed to be in their ground state.\n          Others have demonstrated that changes in geometry can be interpolated as well.\n          For regression, we use kernel ridge regression and Gaussian process regression.\n          Both are kernel-based learning algorithms, a mathematically elegant and systematically non-linear framework for machine learning.\n          Properties are computed at hybrid DFT, GW and G4MP2 levels of theory, and include various energies, polarizability, heat capacity, HOMO and LUMO eigenvalues, depending on dataset.\n        </p>\n        <p>\n          All settings come with a short explanation and/or reference.\n          For a detailed, step-by-step walkthrough of the above, including data and code, see\n        </p>\n        <div style=\"padding: 1ex; margin-top: 1ex; margin-bottom: 1ex; border-style: dotted; border-width: 1pt; border-color: blue; border-radius: 3px;\">\n          M. Rupp: <span style=\"font-style: italic;\">Machine Learning for Quantum Mechanics in a Nutshell</span>, Int. J. Quant. Chem. 115(16): 1058-1073 (2015). <a href=\"http://dx.doi.org/10.1002/qua.24954\">[DOI]</a>\n        </div>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n      </div>\n    </div>\n  </div>\n</div>\n\n<script>\n  qmmlUpdateForm()  // initialize the controls\n  qmmlUpdateProperty()\n  qmmlUpdateRepresentation() \n  qmmlUpdateKernel()\n</script>\n"
+                    "object": "<script>\nvar beaker = bkHelper.getBeakerObject().beakerObj;\n</script>\n<!-- Header -->\n\n<div style=\"font-size: 250%; font-weight: bold;\">Accurate predictions of molecular properties</div>\n\n<!-- <div style=\"font-size: 150%; margin-bottom: 1em;\">A <a href=\"https://nomad-coe.eu/\">NOMAD</a> demonstration by <a href=\"http://mrupp.info/\">Matthias Rupp</a>, 2016.</div> -->\n<div style=\"font-size: 140%; margin-bottom: 1em;\"><a href=\"http://mrupp.info/\">Matthias Rupp</a> <span style=\"font-size: 65%;\">[version 2017-01-25]</span></div>\n\n<div style=\"max-width: 600px;\">\n  This tutorial demonstrates prediction of atomization energies and other properties of small organic molecules at the level of hybrid density functional theory and others, based on nuclear charges and atomic positions only.\n  It is based on\n</div>\n<div style=\"padding: 1ex; margin-top: 1ex; margin-bottom: 1ex; border-style: dotted; border-width: 1pt; border-color: blue; border-radius: 3px; max-width: 600px;\">\n  <p>M. Rupp, A. Tkatchenko, K.-R. Müller, O.A. von Lilienfeld: <span style=\"font-style: italic;\">Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning</span>, Phys. Rev. Lett. 108: 058301 (2012). <a href=\"https://doi.org/10.1103/PhysRevLett.108.058301\">[DOI]</a></p>\n  <!--<p>G. Montavon, M. Rupp, V. Gobre, A. Vazquez-Mayagoitia, K. Hansen, A. Tkatchenko, K.-R. Müller, O.A. von Lilienfeld: <span style=\"font-style: italic;\">Machine learning of molecular electronic properties in chemical compound space</span>, New J. Phys. 15(9): 095003 (2013). <a href=\"http://dx.doi.org/10.1088/1367-2630/15/9/095003\">[DOI]</a></p>-->\n  M. Rupp: <span style=\"font-style: italic;\">Machine Learning for Quantum Mechanics in a Nutshell</span>, Int. J. Quant. Chem. 115(16): 1058-1073 (2015). <a href=\"http://dx.doi.org/10.1002/qua.24954\">[DOI]</a>\n</div>\n<div style=\"max-width: 600px;\">\n  Click on \"Build model!\" below to build a model for atomization energy; click \"Explanation\" for an exposition of the approach; or modify \"Settings\" to produce your own results.\n</div>\n\n<div style=\"padding-top: 2ex; max-width: 600px;\">\n  <span style=\"font-weight: bold;\">Idea: </span> \n  <!--Machine learning can be used to rapidly and accurately predict outcomes of electronic structure calculations.-->\n  To predict ab initio properties across molecules, the problem of solving the molecular Schrödinger equation is mapped onto a nonlinear statistical regression problem.\n  Each electronic structure calculation thus becomes a training point for the regression.\n  The trick is how to represent molecules in an abstract space that supports interpolation.\n</div>\n\n<div style=\"top-margin: 3ex;\">\n  <button type=\"button\" class=\"btn btn-primary\" style=\"margin-top: 2ex;\" onclick=\"trainAndPredict();\">Build model!</button>&nbsp;&nbsp;\n  <button type=\"button\" class=\"btn btn-secondary\" style=\"margin-top: 2ex;\" data-toggle=\"modal\" data-target=\"#qmmldemo-explanation-modal\">Explanation</button>&nbsp;&nbsp;\n  <button type=\"button\" class=\"btn btn-secondary\" style=\"margin-top: 2ex;\" onclick=\"toggle_settings()\">Settings</button>&nbsp;&nbsp;\n  <a target=\"_blank\" href=\"https://nomad-forum.rz-berlin.mpg.de//\"><button type=\"button\" class=\"btn btn-primary\" style=\"margin-top: 2ex;\">Tell us what you think</button></a>    \n</div>\n\n<script>\n  // Shows/hides settings block\n  var toggle_settings = function() {\n    var e = document.getElementById('qmmldemo-hidden-settings-div');\n    if(e.style.display == 'block') { e.style.display = 'none'; } else { e.style.display = 'block'; };\n  }\n  \n  // Adds an option to a dropdown menu\n  function addDropdownChoice(dropdown, value, content) {\n    var el = document.createElement('option');\n    el.value = value;\n    el.innerHTML = content\n    dropdown.appendChild(el);\n  }\n  \n  // Adds an option to a radiobutton selector\n  function addRadiobuttonChoice(radiobutton, value, name, content, checked) {\n    var el = document.createElement('input');\n    el.type = \"radio\";\n    el.value = value;\n    el.name = name;\n    if(checked) { el.checked = \"checked\" }\n\n    var label = document.createElement('label');\n    label.innerHTML = content; label.innerHTML += ' ';\n    label.appendChild(el);\n\n    radiobutton.appendChild(label);\n  }\n  \n  function qmmlUpdateForm() {\n    var ds = document.getElementById(\"qmmldemo_dataset\").value;\n    \n    var dsd = document.getElementById(\"qmmldemo_dataset_description\");\n    var p   = document.getElementById(\"qmmldemo_property\"); p.innerHTML = '';\n    var td  = document.getElementById(\"qmmldemo_traindata\"); td.innerHTML = '';\n\n    switch(ds) {\n      case \"1\":  // PRL 2012\n        dsd.innerHTML = '   M.&nbsp;Rupp <i>et&nbsp;al</i>, Phys. Rev. Lett. 108(5): 058301, 2012  <a href=\"http://dx.doi.org/10.1103/PhysRevLett.108.058301\">DOI</a>';\n\n        addDropdownChoice(p, \"1\", \"atomization energy (DFT/PBE0)\");\n\n        addRadiobuttonChoice(td, \"1000\", \"qmmldemo_traindata_radio\", \"1k\", true );\n        addRadiobuttonChoice(td, \"3000\", \"qmmldemo_traindata_radio\", \"2k\", false);\n        addRadiobuttonChoice(td, \"5000\", \"qmmldemo_traindata_radio\", \"5k\", false);\n        \n        break;\n        \n      case \"2\":  // New J Phys 2013\n        dsd.innerHTML = '   G.&nbsp;Montavon <i>et&nbsp;al</i>, New J. Phys. 15(9): 095003, 2013  <a href=\"http://dx.doi.org/10.1088/1367-2630/15/9/095003\">DOI</a>';\n        \n        addDropdownChoice(p, \"1\", \"atomization energy (DFT/PBE0)\");\n        addDropdownChoice(p, \"2\", \"static polarizability (DFT/PBE0)\");\n        addDropdownChoice(p, \"3\", \"HOMO (DFT/PBE0)\");\n        addDropdownChoice(p, \"4\", \"HOMO (GW)\");\n        addDropdownChoice(p, \"5\", \"LUMO (DFT/PBE0)\");\n        addDropdownChoice(p, \"6\", \"LUMO (GW)\");\n\n        addRadiobuttonChoice(td, \"1000\", \"qmmldemo_traindata_radio\", \"1k\", true );\n        addRadiobuttonChoice(td, \"3000\", \"qmmldemo_traindata_radio\", \"2k\", false);\n        addRadiobuttonChoice(td, \"5000\", \"qmmldemo_traindata_radio\", \"5k\", false);\n        \n        break;\n\n      case \"3\":  //  Sci Data 2014\n        dsd.innerHTML = '   R.&nbsp;Ramakrishnan <i>et&nbsp;al</i>, Sci. Data 1: 140022, 2014  <a href=\"http://dx.doi.org/10.1038/sdata.2014.22\">DOI</a>';\n\n        addDropdownChoice(p, \"1\", \"atomization energy\");\n        \n        addRadiobuttonChoice(td,  \"1000\", \"qmmldemo_traindata_radio\",  \"1k\", true );\n        addRadiobuttonChoice(td,  \"3000\", \"qmmldemo_traindata_radio\",  \"2k\", false);\n        addRadiobuttonChoice(td,  \"5000\", \"qmmldemo_traindata_radio\",  \"5k\", false);\n        addRadiobuttonChoice(td, \"10000\", \"qmmldemo_traindata_radio\", \"10k\", false);\n\n        break;\n        \n      case \"4\":  // Sci Data 2014\n        dsd.innerHTML = '   R.&nbsp;Ramakrishnan <i>et&nbsp;al</i>, Sci. Data 1: 140022, 2014  <a href=\"http://dx.doi.org/10.1038/sdata.2014.22\">DOI</a>';\n\n        addDropdownChoice(p, \"1\", \"zero point vibrational energy (G4MP2)\");\n        addDropdownChoice(p, \"2\", \"internal energy @ 0 K (G4MP2)\");\n        addDropdownChoice(p, \"3\", \"internal energy @ 298.15 K (G4MP2)\");\n        addDropdownChoice(p, \"4\", \"enthalpy @ 298.15 K (G4MP2)\");\n        addDropdownChoice(p, \"5\", \"free energy @ 298.15 K (G4MP2)\");\n        addDropdownChoice(p, \"6\", \"heat capacity @ 298.15 K (DFT/B3LYP)\");\n\n        addRadiobuttonChoice(td, \"1000\", \"qmmldemo_traindata_radio\", \"1k\", true );\n        addRadiobuttonChoice(td, \"3000\", \"qmmldemo_traindata_radio\", \"2k\", false);\n        addRadiobuttonChoice(td, \"5000\", \"qmmldemo_traindata_radio\", \"5k\", false);\n\n        break;\n    }\n  }\n\n  function qmmlUpdateProperty() {\n    var ds  = document.getElementById(\"qmmldemo_dataset\").value;\n    var p   = document.getElementById(\"qmmldemo_property\").value;\n    var pd  = document.getElementById(\"qmmldemo_property_description\");\n\n    switch(ds) {\n      case \"1\": switch(p) {\n        case \"1\": pd.innerHTML = '   atomization energy, DFT/PBE0'; break;\n      }; break;\n      case \"2\": switch(p) {\n        case \"1\": pd.innerHTML = '   atomization energy, DFT/PBE0'; break;\n        case \"2\": pd.innerHTML = '   averaged static polarizability, DFT/PBE0'; break;\n        case \"3\": pd.innerHTML = '   highest occupied molecular orbital, DFT/PBE0'; break;\n        case \"4\": pd.innerHTML = '   highest occupied molecular orbital, GW'; break; \n        case \"5\": pd.innerHTML = '   lowest unoccupied molecular orbital, DFT/PBE0'; break; \n        case \"6\": pd.innerHTML = '   lowest unoccupied molecular orbital, GW'; break; \n      }; break;\n      case \"3\": switch(p) {\n        case \"1\": pd.innerHTML = '   atomization energy, DFT/B3LYP/6-31G(2df,p)'; break;\n      }; break;\n      case \"4\": switch(p) {\n        case \"1\": pd.innerHTML = '   zero point vibrational energy, G4MP2'; break;\n        case \"2\": pd.innerHTML = '   internal energy at 0 K, G4MP2'; break;\n        case \"3\": pd.innerHTML = '   internal energy at 298.15 K, G4MP2'; break;\n        case \"4\": pd.innerHTML = '   enthalpy at 298.15 K, G4MP2'; break;\n        case \"5\": pd.innerHTML = '   free energy at 298.15 K, G4MP2'; break;\n        case \"6\": pd.innerHTML = '   heat capacity at 298.15 K, DFT/B3LYP'; break;\n      }; break;\n    }\n  }\n\n  function qmmlUpdateRepresentation() {\n    var r   = document.getElementById(\"qmmldemo_representation\").value;\n    var rd  = document.getElementById(\"qmmldemo_representation_description\");\n    switch(r) {\n      case \"1\": rd.innerHTML = '   M.&nbsp;Rupp <i>et&nbsp;al</i>, Phys. Rev. Lett. 108(5): 058301, 2012  <a href=\"http://dx.doi.org/10.1103/PhysRevLett.108.058301\">DOI</a>'; break;\n      case \"2\": rd.innerHTML = '   M.&nbsp;Rupp <i>et&nbsp;al</i>, Phys. Rev. Lett. 108(5): 058301, 2012  <a href=\"http://dx.doi.org/10.1103/PhysRevLett.108.058301\">DOI</a>'; break;\n      case \"3\": rd.innerHTML = '   K.&nbsp;Hansen <i>et&nbsp;al</i>, J. Phys. Chem. Lett. 6: 2326, 2015  <a href=\"http://dx.doi.org/10.1021/acs.jpclett.5b00831\">DOI</a>'; break;\n    }\n  }\n  \n  function qmmlUpdateKernel() {\n    var k  = document.getElementById(\"qmmldemo_kernel\").value;\n    var kd = document.getElementById(\"qmmldemo_kernel_description\");\n    switch(k) {\n      case \"1\": kd.innerHTML = \"   linear basis functions, no free parameters\"; break;\n      case \"2\": kd.innerHTML = \"   Gaussian basis functions, length scale free parameter\"; break;\n      case \"3\": kd.innerHTML = \"   exponential basis functions, length scale free parameter\"; break;\n    }\n  }\n  \n  function trainAndPredict() {\n    beaker.ctrl_ds   = document.getElementById(\"qmmldemo_dataset\").value;\n    beaker.ctrl_pty  = document.getElementById(\"qmmldemo_property\").value;\n    beaker.ctrl_repr = document.getElementById(\"qmmldemo_representation\").value;\n    beaker.ctrl_k    = document.getElementById(\"qmmldemo_kernel\").value;\n    beaker.ctrl_n    = document.querySelector('input[name=\"qmmldemo_traindata_radio\"]:checked').value;\n    beaker.evaluate(\"trainAndPredictCell\");\n  }\n</script>\n\n<style type=\"text/css\">\n  \n  .qmmldemo_table th { font-weight: bold; padding-right: 2ex; }\n  .qmmldemo_table td input { margin-right: 1ex; }\n  \n</style>\n\n<!-- Controls area -->\n\n<div class=\"qmmldemo_control\" id=\"qmmldemo-hidden-settings-div\" style=\"display: none;\">\n  <p style=\"height: 2ex;\"></p>\n  <table class=\"qmmldemo_table\">\n    \n    <tbody><tr><th>Dataset:</th>\n      <td>\n        <select id=\"qmmldemo_dataset\" onchange=\"qmmlUpdateForm()\">\n          <option value=\"1\" selected=\"\">7k small organic molecules (GDB7-12)</option>\n          <option value=\"2\">7k small organic molecules (GDB7-13)</option>\n          <!-- <option value=\"3\">134k small organic molecules (GDB9-14)</option> -->\n          <option value=\"4\">6k C7H20O2 isomers (GDB9-14)</option>\n        </select>\n      </td>\n      <td id=\"qmmldemo_dataset_description\" style=\"white-space: pre;\">   M.&nbsp;Rupp <i>et&nbsp;al</i>, Phys. Rev. Lett. 108(5): 058301, 2012  <a href=\"http://dx.doi.org/10.1103/PhysRevLett.108.058301\">DOI</a></td>\n    </tr>\n    \n    <tr>\n      <th>Property:</th>\n      <td><select id=\"qmmldemo_property\" onchange=\"qmmlUpdateProperty()\"><option value=\"1\">atomization energy (DFT/PBE0)</option></select></td>\n      <td id=\"qmmldemo_property_description\" style=\"white-space: pre;\">   atomization energy, DFT/PBE0</td>\n    </tr>\n    \n    <tr>\n      <th>Representation:</th>\n      <td><select id=\"qmmldemo_representation\" onchange=\"qmmlUpdateRepresentation()\">\n        <option value=\"1\">sorted Coulomb matrix</option>\n        <option value=\"2\">diagonalized Coulomb matrix</option>\n        <!-- <option value=\"3\">bag of bonds</option> -->\n       </select></td>\n      <td id=\"qmmldemo_representation_description\" style=\"white-space: pre;\">   M.&nbsp;Rupp <i>et&nbsp;al</i>, Phys. Rev. Lett. 108(5): 058301, 2012  <a href=\"http://dx.doi.org/10.1103/PhysRevLett.108.058301\">DOI</a></td>\n    </tr>\n    \n    <tr>\n      <th>Kernel:</th>\n      <td><select id=\"qmmldemo_kernel\" onchange=\"qmmlUpdateKernel()\">\n        <option value=\"1\">linear kernel</option>\n        <option value=\"2\">Gaussian kernel</option>\n        <option value=\"3\">Laplacian kernel</option>         \n      </select></td>\n      <td id=\"qmmldemo_kernel_description\" style=\"white-space: pre;\">   linear basis functions, no free parameters</td>\n    </tr>\n    \n    <tr>\n      <th>Training data:</th>\n      <td><form id=\"qmmldemo_traindata\"><label>1k <input value=\"1000\" name=\"qmmldemo_traindata_radio\" type=\"radio\"></label><label>2k <input value=\"3000\" name=\"qmmldemo_traindata_radio\" type=\"radio\"></label><label>5k <input value=\"5000\" name=\"qmmldemo_traindata_radio\" type=\"radio\"></label></form></td>\n      <td style=\"white-space: pre;\">   number of training examples</td>\n    </tr>\n  </tbody></table>\n</div>\n\n<!-- Explanation -->\n<div class=\"modal fade\" id=\"qmmldemo-explanation-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"qmmldemo-explanation-modal-label\" style=\"display: none;\">\n  <div class=\"modal-dialog modal-lg\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"qmmldemo-explanation-modal-label\">Explanation</h4>\n      </div>\n      <div class=\"modal-body qmmldemo_explanation\">\n        <p>\n          In this tutorial we demonstrate fast and accurate prediction of ab initio properties of small organic molecules.\n        </p>\n        <p>\n          The idea is that computing the same property for many similar molecules entails a high degree of redundancy:\n          The same deterministic numeric procedure is carried out repeatedly for correlated inputs, yielding correlated outputs.\n          Consequently, machine learning can be used to interpolate between reference calculations, exploiting this redundancy.\n        </p>\n        <p>\n          In such models, the problem of repeatedly solving the electronic Schrödinger equation is mapped onto a non-linear regression problem.\n          The key challenge is to define a space (or basis set) in which each reference calculation is a point and that enables efficient interpolation.\n          If successful, the interpolation / prediction error decreases with the number of reference calculations (training data):\n        </p>\n        <img src=\"http://iopscience.iop.org/1367-2630/15/9/095003/downloadHRFigure/figure/nj459085f2\" alt=\"learning curve\" style=\"margin: 2ex; width: 60%; padding-left: 3ex;\">\n        <p>\n          Here, MAE = mean absolute error over a validation set; predictions are for E = atomization energy, α = polarizability.\n          The machine learning predictions themselves have negligible computational cost (on the order of milliseconds) compared to electronic structure calculations.\n        </p>\n        <p>\n          This tutorial aims at demonstrating interpolation across changes in chemistry.\n          Molecules are therefore assumed to be in their ground state.\n          Others have demonstrated that changes in geometry can be interpolated as well.\n          For regression, we use kernel ridge regression and Gaussian process regression.\n          Both are kernel-based learning algorithms, a mathematically elegant and systematically non-linear framework for machine learning.\n          Properties are computed at hybrid DFT, GW and G4MP2 levels of theory, and include various energies, polarizability, heat capacity, HOMO and LUMO eigenvalues, depending on dataset.\n        </p>\n        <p>\n          All settings come with a short explanation and/or reference.\n          For a detailed, step-by-step walkthrough of the above, including data and code, see\n        </p>\n        <div style=\"padding: 1ex; margin-top: 1ex; margin-bottom: 1ex; border-style: dotted; border-width: 1pt; border-color: blue; border-radius: 3px;\">\n          M. Rupp: <span style=\"font-style: italic;\">Machine Learning for Quantum Mechanics in a Nutshell</span>, Int. J. Quant. Chem. 115(16): 1058-1073 (2015). <a href=\"http://dx.doi.org/10.1002/qua.24954\">[DOI]</a>\n        </div>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n      </div>\n    </div>\n  </div>\n</div>\n\n<script>\n  qmmlUpdateForm()  // initialize the controls\n  qmmlUpdateProperty()\n  qmmlUpdateRepresentation() \n  qmmlUpdateKernel()\n</script>\n"
                 },
                 "elapsedTime": 0
             },
diff --git a/beaker-notebooks/clusterX.bkr b/beaker-notebooks/clusterX.bkr
index 802f56d730c1f40ce8f15f5b9730f272e5410fdb..af67296f8daf651dadc949669d5f90d4d90afd4e 100644
--- a/beaker-notebooks/clusterX.bkr
+++ b/beaker-notebooks/clusterX.bkr
@@ -289,7 +289,7 @@
                     "  Settings",
                     "</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;",
                     "",
-                    "<a target=\"_blank\" href=\"http://forum.analytics-toolkit.nomad-coe.eu/\" class=\"btn btn-primary\"> Tell us what you think</a>",
+                    "<a target=\"_blank\" href=\"https://nomad-forum.rz-berlin.mpg.de//\" class=\"btn btn-primary\"> Tell us what you think</a>",
                     ""
                 ],
                 "hidden": true
@@ -299,7 +299,7 @@
                 "result": {
                     "type": "BeakerDisplay",
                     "innertype": "Html",
-                    "object": "<script>\nvar beaker = bkHelper.getBeakerObject().beakerObj;\n</script>\n<script>\n  var run_lasso = function() {\n  $(\"#lasso_result_button\").removeClass(\"active\").addClass(\"disabled\");\n  myFunc();\n  //getFeatures();\n  //getOperators();\n  //beaker.max_dim = $(\"#lasso_max_dim_selector\").val();\n  //beaker.structures_diff = $(\"#lasso_structures_diff\").val();\n  //beaker.n_comb = $(\"#n_comb\").val();\n  //beaker.n_sis = $(\"#n_sis\").val();\n  //beaker.units = $(\"#units_select\").val();\n  beaker.evaluate(\"calc_cell1\"); // evaluate cells with tag \"lasso_cell\"\n  beaker.evaluate(\"calc_cell2\"); // evaluate cells with tag \"lasso_cell\"\n  beaker.evaluate(\"calc_cell3\"); // evaluate cells with tag \"lasso_cell\"\n  beaker.evaluate(\"calc_cell4\"); // evaluate cells with tag \"lasso_cell\"\n  beaker.evaluate(\"calc_cell5\"); // evaluate cells with tag \"lasso_cell\"\n  beaker.evaluate(\"calc_cell6\"); // evaluate cells with tag \"lasso_cell\"\n  // view_result()\n  };\n\n  var reset_lasso = function(){\n  beaker.evaluate(\"lasso-settings-cell\");\n  var e = document.getElementById('lasso-hidden-settings-div');\n  var b = document.getElementById('lasso-hidden-settings-button');\n  e.style.display = 'block';\n  b.style.display = 'inline';\n  //beaker.evaluate(\"calc_cell1\");\n  var c = document.getElementById('lasso-hidden-out');\n  c.style.display = 'none';\n  //beaker.evaluate(\"calc_cell1\");\n  };\n\n  var getFeatures = function() {\n  beaker.selected_feature_list = [];\n  $('#lasso_features_select input:checkbox').each(function () {\n  if(this.checked )\n  beaker.selected_feature_list.push(this.value);\n  });\n  };\n\n  var getOperators = function() {\n  beaker.allowed_operations = [];\n  $('#lasso_operators_select input:checkbox').each(function () {\n  if(this.checked )\n  beaker.allowed_operations.push(this.value);\n  });\n  };  \n\n  var toggle_settings = function(){\n  var e = document.getElementById('lasso-hidden-settings-div');\n  var b = document.getElementById('lasso-hidden-settings-button');\n  if(e.style.display == 'block'){\n  e.style.display = 'none';\n  b.style.display = 'none';\n  }\n  else{\n  e.style.display = 'block';\n  b.style.display = 'inline';\n  }\n  };\n  beaker.view_result = function(result_link) {\n  //   beaker.evaluate(\"lasso_viewer_result\").then(function(x) {\n  $(\"#lasso_result_button\").attr(\"href\", result_link);\n  //   }); \n  $(\"#lasso_result_button\").removeClass(\"disabled\").addClass(\"active\");\n  }\n\n\n  var myFunc = function() {\n  beaker.cradius = document.getElementsByName(\"cradius\")[0].value;\n  beaker.cpoints = document.getElementById(\"cpoints\").value;\n  beaker.msteps = document.getElementsByName(\"msteps\")[0].value;\n  beaker.mtemp = document.getElementsByName(\"mtemp\")[0].value;\n  beaker.nsub = document.getElementsByName(\"nsub\")[0].value;\n  //beaker.evaluate(\"\");\n  };\n</script>\n\n<style type=\"text/css\">\n  .lasso_instructions{\n  font-size: 15px;\n  } \n</style>\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" data-toggle=\"modal\" data-target=\"#lasso-motivation-modal\">\n  Background\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<!-- Modal -->\n<div class=\"modal fade\" id=\"lasso-motivation-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"lasso-motivation-modal-label\" style=\"display: none;\">\n  <div class=\"modal-dialog modal-lg\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"lasso-motivation-modal-label\">Background</h4>\n      </div>\n      <div class=\"modal-body lasso_instructions\">\n\t<p>Substitutional alloys, such as, <i>e.g.</i>, Si<sub>1-x</sub>Ge<sub>x</sub>, present a multitude of atomic arrangements, or configurations. Thus, it is impossible to perform numerically costly <i>ab-initio</i> calculations for every of them. The cluster-expansion (CE) technique allows for building numerically efficient models to predict the energy E of the alloy by exploiting the unique dependence of E on the configuration (<i>i.e.</i> the specific arrangement of Ge atoms in the crystal). It relies on a set of <i>ab-initio</i> calculations obtained by density-functional theory and has been applied to describe stable phases of bulk and surface alloys</p>\n\n\t<p>With this technique, the energy can be parametrized in terms of clusters (a set of crystal sites). The energy contribution of every of them is termed effective-cluster interaction (ECI). Thus, for an arbitrary configuration of the alloy, the clusters formed by the substitutional species are found and the sum of the respective ECIs gives a prediction of the energy for the given configuration. Each cluster is defined by a specific number and arrangement of crystal sites (empty cluster, 1-point clusters, ..., n-point clusters). </p>\n\t<p>\n\t  Through a cross-validation procedure, the optimal set of clusters to represent the energy of the alloy are found. This set is then used to build a model which allows quick and accurate predictions of the energy for arbitrary configurations. Using this model, millions of structures can be sampled, and thus the ground state configurations can be found. The knowledge of the stable structures is of fundamental importance for the understanding of the physical properties of the alloys. \n\t</p>\n\t<p>\n\t  The tool makes use of the python package <tt>CELL</tt> (Cluster Expansion for large parent ceLLs). An application of <tt>CELL</tt>  for the prediction of stable phases of complex thermoelectric alloys, can be found in:\n\t  </p><div style=\"padding: 1ex; margin-top: 1ex; margin-bottom: 1ex; border-style: dotted; border-width: 1pt; border-color: blue; border-radius: 3px;\">\n\t    M. Troppenz, S. Rigamonti, and C. Draxl: <span style=\"font-style: italic;\">Predicting Ground-State Configurations and Electronic Properties of the\n\t      Thermoelectric Clathrates Ba<sub>8</sub>Al<sub>x</sub>Si<sub>46-x</sub> and Sr<sub>8</sub>Al<sub>x</sub>Si<sub>46-x</sub></span>, accepted for publication in Chemistry of Materials (2017).\n\t  </div>\n\t<p></p>\n\t\n\t<p><tt>CELL</tt> uses the <b>corrdump</b> utility of ATAT for the generation of clusters and the construction of the correlation matrix (A. van de Walle, CALPHAD: Comput. Coupling Phase Diagrams Thermochem. 33, 266-278 (2009))</p>\n\n\t<p>References:</p>\n        <ol>\n          <li>J. M. Sanchez, F. Ducastelle, and D. Gratias, Phys. A 128, 334-350 (1984).</li>\n          <li>A. van de Walle and G. Ceder, J. Phase Equilib. 23, 348-359 (2002).</li>\n        </ol>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n\t<!--         <button type=\"button\" class=\"btn btn-primary\">Save changes</button> -->\n      </div>\n    </div>\n  </div>\n</div>\n\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" data-toggle=\"modal\" data-target=\"#lasso-instructions-modal\">\n  Instructions\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<!-- Modal -->\n<div class=\"modal fade\" id=\"lasso-instructions-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"lasso-instructions-modal-label\" style=\"display: none;\">\n  <div class=\"modal-dialog\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"lasso-instructions-modal-label\">Instructions</h4>\n      </div>\n      <div class=\"modal-body lasso_instructions\">\n\t<p>In this example, you can perform a cluster expansion (CE) of the binary alloy Si<sub>x</sub>Ge<sub>1-x</sub>. By finding the optimal set of cluster, a model is built which allows numerically efficient predictions of the energy of any configuration (i.e. atomic arrangement) of the substitutional species Ge. Finally, you can perform a configurational sampling to predict the most stable configurations of the alloy.</p>\n\t<p>\n\t  The figure below shows the 16-atoms supercell used to perform the study. An exemplary\n\t  </p><figure>\n\t    <img src=\"/user/struc.png\" alt=\"Supercell\" style=\"width:354px;height:304px;\">\n\t  </figure>\n \t<p></p>\n\n\t<p>The optimal set of clusters is selected out of a large pool of clusters through a cross validation procedure.</p>\n\n\t<p>By clicking <b>Settings</b> you can set the parameters determining the size and charachter of the pool of clusters. You can also set the parameters affecting the configurational sampling. A more detailed explanation of the different settings can be found below: \n          </p><ul>\n            <li>Maximum radius (Angstrom): Maximum distance between any two crystal sites composing a cluster in the pool of clusters. </li>\n            <li>Maximum number of points: Maximum number of points (i.e. crystal sites) composing a cluster in the pool of clusters.</li>\n            <li>Number of sampling steps: Number of configurations sampled at every composition x.</li>\n            <li>Temperature (K): Temperature of the simulated annealing procedure used for sampling the configuration space. The sampling is based on a canonical Metropolis algorithm.</li>\n            <li>Number of concentrations to be sampled: Number of values of x (in the formula Si<sub>x</sub>Ge<sub>1-x</sub>) for which samplings are performed.</li>\n          </ul>    \n          \n\t<p></p>\n\t\n\t\n\t<p>        After the preferred settings have been adjusted, click <b>RUN</b> for performing the calculations. </p>\n\n\tDuring the run, a brief summary is printed out below the <b>RUN</b> button. At the end of the run: \n\t<ul>\n\t  <li> the cluster optimization procedure is plotted. Both the training-RMSE and cv-RMSE error are depicted. A red circle in the plot indicates the set of clusters leading to the lowest cv-RMSE. This is used to build the model to predict energies in the configurational sampling.</li>\n\t  <li> A plot of the energy of mixing versus composition is displayed. Both the <i>ab-initio</i>, fitted, sampled and lowest energy found are depicted.</li>\n\t</ul>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n\t<!--         <button type=\"button\" class=\"btn btn-primary\">Save changes</button> -->\n      </div>\n    </div>\n  </div>\n</div>\n\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" onclick=\"toggle_settings()\">\n  Settings\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<a target=\"_blank\" href=\"http://forum.analytics-toolkit.nomad-coe.eu/\" class=\"btn btn-primary\"> Tell us what you think</a>\n"
+                    "object": "<script>\nvar beaker = bkHelper.getBeakerObject().beakerObj;\n</script>\n<script>\n  var run_lasso = function() {\n  $(\"#lasso_result_button\").removeClass(\"active\").addClass(\"disabled\");\n  myFunc();\n  //getFeatures();\n  //getOperators();\n  //beaker.max_dim = $(\"#lasso_max_dim_selector\").val();\n  //beaker.structures_diff = $(\"#lasso_structures_diff\").val();\n  //beaker.n_comb = $(\"#n_comb\").val();\n  //beaker.n_sis = $(\"#n_sis\").val();\n  //beaker.units = $(\"#units_select\").val();\n  beaker.evaluate(\"calc_cell1\"); // evaluate cells with tag \"lasso_cell\"\n  beaker.evaluate(\"calc_cell2\"); // evaluate cells with tag \"lasso_cell\"\n  beaker.evaluate(\"calc_cell3\"); // evaluate cells with tag \"lasso_cell\"\n  beaker.evaluate(\"calc_cell4\"); // evaluate cells with tag \"lasso_cell\"\n  beaker.evaluate(\"calc_cell5\"); // evaluate cells with tag \"lasso_cell\"\n  beaker.evaluate(\"calc_cell6\"); // evaluate cells with tag \"lasso_cell\"\n  // view_result()\n  };\n\n  var reset_lasso = function(){\n  beaker.evaluate(\"lasso-settings-cell\");\n  var e = document.getElementById('lasso-hidden-settings-div');\n  var b = document.getElementById('lasso-hidden-settings-button');\n  e.style.display = 'block';\n  b.style.display = 'inline';\n  //beaker.evaluate(\"calc_cell1\");\n  var c = document.getElementById('lasso-hidden-out');\n  c.style.display = 'none';\n  //beaker.evaluate(\"calc_cell1\");\n  };\n\n  var getFeatures = function() {\n  beaker.selected_feature_list = [];\n  $('#lasso_features_select input:checkbox').each(function () {\n  if(this.checked )\n  beaker.selected_feature_list.push(this.value);\n  });\n  };\n\n  var getOperators = function() {\n  beaker.allowed_operations = [];\n  $('#lasso_operators_select input:checkbox').each(function () {\n  if(this.checked )\n  beaker.allowed_operations.push(this.value);\n  });\n  };  \n\n  var toggle_settings = function(){\n  var e = document.getElementById('lasso-hidden-settings-div');\n  var b = document.getElementById('lasso-hidden-settings-button');\n  if(e.style.display == 'block'){\n  e.style.display = 'none';\n  b.style.display = 'none';\n  }\n  else{\n  e.style.display = 'block';\n  b.style.display = 'inline';\n  }\n  };\n  beaker.view_result = function(result_link) {\n  //   beaker.evaluate(\"lasso_viewer_result\").then(function(x) {\n  $(\"#lasso_result_button\").attr(\"href\", result_link);\n  //   }); \n  $(\"#lasso_result_button\").removeClass(\"disabled\").addClass(\"active\");\n  }\n\n\n  var myFunc = function() {\n  beaker.cradius = document.getElementsByName(\"cradius\")[0].value;\n  beaker.cpoints = document.getElementById(\"cpoints\").value;\n  beaker.msteps = document.getElementsByName(\"msteps\")[0].value;\n  beaker.mtemp = document.getElementsByName(\"mtemp\")[0].value;\n  beaker.nsub = document.getElementsByName(\"nsub\")[0].value;\n  //beaker.evaluate(\"\");\n  };\n</script>\n\n<style type=\"text/css\">\n  .lasso_instructions{\n  font-size: 15px;\n  } \n</style>\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" data-toggle=\"modal\" data-target=\"#lasso-motivation-modal\">\n  Background\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<!-- Modal -->\n<div class=\"modal fade\" id=\"lasso-motivation-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"lasso-motivation-modal-label\" style=\"display: none;\">\n  <div class=\"modal-dialog modal-lg\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"lasso-motivation-modal-label\">Background</h4>\n      </div>\n      <div class=\"modal-body lasso_instructions\">\n\t<p>Substitutional alloys, such as, <i>e.g.</i>, Si<sub>1-x</sub>Ge<sub>x</sub>, present a multitude of atomic arrangements, or configurations. Thus, it is impossible to perform numerically costly <i>ab-initio</i> calculations for every of them. The cluster-expansion (CE) technique allows for building numerically efficient models to predict the energy E of the alloy by exploiting the unique dependence of E on the configuration (<i>i.e.</i> the specific arrangement of Ge atoms in the crystal). It relies on a set of <i>ab-initio</i> calculations obtained by density-functional theory and has been applied to describe stable phases of bulk and surface alloys</p>\n\n\t<p>With this technique, the energy can be parametrized in terms of clusters (a set of crystal sites). The energy contribution of every of them is termed effective-cluster interaction (ECI). Thus, for an arbitrary configuration of the alloy, the clusters formed by the substitutional species are found and the sum of the respective ECIs gives a prediction of the energy for the given configuration. Each cluster is defined by a specific number and arrangement of crystal sites (empty cluster, 1-point clusters, ..., n-point clusters). </p>\n\t<p>\n\t  Through a cross-validation procedure, the optimal set of clusters to represent the energy of the alloy are found. This set is then used to build a model which allows quick and accurate predictions of the energy for arbitrary configurations. Using this model, millions of structures can be sampled, and thus the ground state configurations can be found. The knowledge of the stable structures is of fundamental importance for the understanding of the physical properties of the alloys. \n\t</p>\n\t<p>\n\t  The tool makes use of the python package <tt>CELL</tt> (Cluster Expansion for large parent ceLLs). An application of <tt>CELL</tt>  for the prediction of stable phases of complex thermoelectric alloys, can be found in:\n\t  </p><div style=\"padding: 1ex; margin-top: 1ex; margin-bottom: 1ex; border-style: dotted; border-width: 1pt; border-color: blue; border-radius: 3px;\">\n\t    M. Troppenz, S. Rigamonti, and C. Draxl: <span style=\"font-style: italic;\">Predicting Ground-State Configurations and Electronic Properties of the\n\t      Thermoelectric Clathrates Ba<sub>8</sub>Al<sub>x</sub>Si<sub>46-x</sub> and Sr<sub>8</sub>Al<sub>x</sub>Si<sub>46-x</sub></span>, accepted for publication in Chemistry of Materials (2017).\n\t  </div>\n\t<p></p>\n\t\n\t<p><tt>CELL</tt> uses the <b>corrdump</b> utility of ATAT for the generation of clusters and the construction of the correlation matrix (A. van de Walle, CALPHAD: Comput. Coupling Phase Diagrams Thermochem. 33, 266-278 (2009))</p>\n\n\t<p>References:</p>\n        <ol>\n          <li>J. M. Sanchez, F. Ducastelle, and D. Gratias, Phys. A 128, 334-350 (1984).</li>\n          <li>A. van de Walle and G. Ceder, J. Phase Equilib. 23, 348-359 (2002).</li>\n        </ol>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n\t<!--         <button type=\"button\" class=\"btn btn-primary\">Save changes</button> -->\n      </div>\n    </div>\n  </div>\n</div>\n\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" data-toggle=\"modal\" data-target=\"#lasso-instructions-modal\">\n  Instructions\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<!-- Modal -->\n<div class=\"modal fade\" id=\"lasso-instructions-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"lasso-instructions-modal-label\" style=\"display: none;\">\n  <div class=\"modal-dialog\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"lasso-instructions-modal-label\">Instructions</h4>\n      </div>\n      <div class=\"modal-body lasso_instructions\">\n\t<p>In this example, you can perform a cluster expansion (CE) of the binary alloy Si<sub>x</sub>Ge<sub>1-x</sub>. By finding the optimal set of cluster, a model is built which allows numerically efficient predictions of the energy of any configuration (i.e. atomic arrangement) of the substitutional species Ge. Finally, you can perform a configurational sampling to predict the most stable configurations of the alloy.</p>\n\t<p>\n\t  The figure below shows the 16-atoms supercell used to perform the study. An exemplary\n\t  </p><figure>\n\t    <img src=\"/user/struc.png\" alt=\"Supercell\" style=\"width:354px;height:304px;\">\n\t  </figure>\n \t<p></p>\n\n\t<p>The optimal set of clusters is selected out of a large pool of clusters through a cross validation procedure.</p>\n\n\t<p>By clicking <b>Settings</b> you can set the parameters determining the size and charachter of the pool of clusters. You can also set the parameters affecting the configurational sampling. A more detailed explanation of the different settings can be found below: \n          </p><ul>\n            <li>Maximum radius (Angstrom): Maximum distance between any two crystal sites composing a cluster in the pool of clusters. </li>\n            <li>Maximum number of points: Maximum number of points (i.e. crystal sites) composing a cluster in the pool of clusters.</li>\n            <li>Number of sampling steps: Number of configurations sampled at every composition x.</li>\n            <li>Temperature (K): Temperature of the simulated annealing procedure used for sampling the configuration space. The sampling is based on a canonical Metropolis algorithm.</li>\n            <li>Number of concentrations to be sampled: Number of values of x (in the formula Si<sub>x</sub>Ge<sub>1-x</sub>) for which samplings are performed.</li>\n          </ul>    \n          \n\t<p></p>\n\t\n\t\n\t<p>        After the preferred settings have been adjusted, click <b>RUN</b> for performing the calculations. </p>\n\n\tDuring the run, a brief summary is printed out below the <b>RUN</b> button. At the end of the run: \n\t<ul>\n\t  <li> the cluster optimization procedure is plotted. Both the training-RMSE and cv-RMSE error are depicted. A red circle in the plot indicates the set of clusters leading to the lowest cv-RMSE. This is used to build the model to predict energies in the configurational sampling.</li>\n\t  <li> A plot of the energy of mixing versus composition is displayed. Both the <i>ab-initio</i>, fitted, sampled and lowest energy found are depicted.</li>\n\t</ul>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n\t<!--         <button type=\"button\" class=\"btn btn-primary\">Save changes</button> -->\n      </div>\n    </div>\n  </div>\n</div>\n\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" onclick=\"toggle_settings()\">\n  Settings\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<a target=\"_blank\" href=\"https://nomad-forum.rz-berlin.mpg.de//\" class=\"btn btn-primary\"> Tell us what you think</a>\n"
                 },
                 "selectedType": "BeakerDisplay",
                 "elapsedTime": 0,
diff --git a/beaker-notebooks/phasediagram.bkr b/beaker-notebooks/phasediagram.bkr
index ec63426555524c7ccdef6c77e3025e9ab5068ad1..33a63c4345be0a3eba7591529d43c4e195ad680a 100644
--- a/beaker-notebooks/phasediagram.bkr
+++ b/beaker-notebooks/phasediagram.bkr
@@ -62,7 +62,7 @@
             "evaluator": "HTML",
             "input": {
                 "body": [
-                    "<a target=\"_blank\" href=\"http://forum.analytics-toolkit.nomad-coe.eu/\" class=\"btn btn-primary\"> Send your feedback to the analytics-toolkit forum</a><h2> Your comments are invaluable in helping us provide a user friendly experience for all! </h2>"
+                    "<a target=\"_blank\" href=\"https://nomad-forum.rz-berlin.mpg.de//\" class=\"btn btn-primary\"> Send your feedback to the analytics-toolkit forum</a><h2> Your comments are invaluable in helping us provide a user friendly experience for all! </h2>"
                 ]
             },
             "output": {
@@ -70,7 +70,7 @@
                 "result": {
                     "type": "BeakerDisplay",
                     "innertype": "Html",
-                    "object": "<script>\nvar beaker = bkHelper.getBeakerObject().beakerObj;\n</script>\n<a target=\"_blank\" href=\"http://forum.analytics-toolkit.nomad-coe.eu/\" class=\"btn btn-primary\"> Send your feedback to the analytics-toolkit forum</a><h2> Your comments are invaluable in helping us provide a user friendly experience for all! </h2>"
+                    "object": "<script>\nvar beaker = bkHelper.getBeakerObject().beakerObj;\n</script>\n<a target=\"_blank\" href=\"https://nomad-forum.rz-berlin.mpg.de//\" class=\"btn btn-primary\"> Send your feedback to the analytics-toolkit forum</a><h2> Your comments are invaluable in helping us provide a user friendly experience for all! </h2>"
                 },
                 "selectedType": "BeakerDisplay",
                 "elapsedTime": 0,
diff --git a/beaker-notebooks/sis_cscl.bkr b/beaker-notebooks/sis_cscl.bkr
index 4d3cea646e39b70261b2ad95432a65ba5520c0af..329cefe908b78b6929820a4a9b04aac9c584227e 100644
--- a/beaker-notebooks/sis_cscl.bkr
+++ b/beaker-notebooks/sis_cscl.bkr
@@ -246,7 +246,7 @@
                     " Settings",
                     "</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;",
                     "",
-                    "<a target=\"_blank\" href=\"http://forum.analytics-toolkit.nomad-coe.eu/\" class=\"btn btn-primary\"> Tell us what you think</a>",
+                    "<a target=\"_blank\" href=\"https://nomad-forum.rz-berlin.mpg.de//\" class=\"btn btn-primary\"> Tell us what you think</a>",
                     "",
                     ""
                 ],
@@ -257,7 +257,7 @@
                 "result": {
                     "type": "BeakerDisplay",
                     "innertype": "Html",
-                    "object": "<script>\nvar beaker = bkHelper.getBeakerObject().beakerObj;\n</script>\n<script>\nvar run_lasso = function() {\n  $(\"#lasso_result_button\").removeClass(\"active\").addClass(\"disabled\");\n  getFeatures();\n  getOperators();\n  beaker.max_dim = $(\"#lasso_max_dim_selector\").val();\n  beaker.structures_diff = $(\"#lasso_structures_diff\").val();\n  beaker.n_comb = $(\"#n_comb\").val();\n  beaker.n_sis = $(\"#n_sis\").val();\n  beaker.units = $(\"#units_select\").val();\n  beaker.evaluate(\"calc_cell\"); // evaluate cells with tag \"lasso_cell\"\n // view_result()\n};\nvar reset_lasso = function(){\n  beaker.evaluate(\"lasso-settings-cell\");\n  var e = document.getElementById('lasso-hidden-settings-div');\n  var b = document.getElementById('lasso-hidden-settings-button');\n  e.style.display = 'block';\n  b.style.display = 'inline';\n};\nvar getFeatures = function() {\n    beaker.selected_feature_list = [];\n    $('#lasso_features_select input:checkbox').each(function () {\n        if(this.checked )\n          beaker.selected_feature_list.push(this.value);\n    });\n};\nvar getOperators = function() {\n    beaker.allowed_operations = [];\n    $('#lasso_operators_select input:checkbox').each(function () {\n        if(this.checked )\n          beaker.allowed_operations.push(this.value);\n    });\n};  \n\nvar toggle_settings = function(){\n  var e = document.getElementById('lasso-hidden-settings-div');\n  var b = document.getElementById('lasso-hidden-settings-button');\n  if(e.style.display == 'block'){\n    e.style.display = 'none';\n    b.style.display = 'none';\n  }\n  else{\n    e.style.display = 'block';\n    b.style.display = 'inline';\n  }\n};\nbeaker.view_result = function(result_link) {\n//   beaker.evaluate(\"lasso_viewer_result\").then(function(x) {\n    $(\"#lasso_result_button\").attr(\"href\", result_link);\n//   }); \n  $(\"#lasso_result_button\").removeClass(\"disabled\").addClass(\"active\");\n}\n\n\n\n\n\n\n\n</script>\n<style type=\"text/css\">\n .lasso_instructions{\n    font-size: 15px;\n  } \n</style>\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" data-toggle=\"modal\" data-target=\"#lasso-motivation-modal\">\n Background\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<!-- Modal -->\n<div class=\"modal fade\" id=\"lasso-motivation-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"lasso-motivation-modal-label\">\n  <div class=\"modal-dialog modal-lg\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"lasso-motivation-modal-label\">Background</h4>\n      </div>\n      <div class=\"modal-body lasso_instructions\">\n        <p>We present a tool for predicting the crystal structure of octet binary compounds, by using a set of descriptive parameters (a descriptor) based on free-atom data of the atomic species constituting the binary material. This task is similar to the one presented in the tutorial <a href=\"http://analytics-toolkit.nomad-coe.eu/tutorial-LASSO_L0\">\"[Predicting energy differences between different crystal structures I: large feature space]\"</a>\". In contrast to that tutorial, here we apply a newly developed method: Sure Independent Screening - Sparse Approximation (SIS-SA), that allows to find an optimal descriptor in a huge feature space containing billions of features.\nThe method is described in:\n</p><div style=\"padding: 1ex; margin-top: 1ex; margin-bottom: 1ex; border-style: dotted; border-width: 1pt; border-color: blue; border-radius: 3px;\">\nR. Ouyang, E. Ahmetcik, L. M. Ghiringhelli, and M. Scheffler: <span style=\"font-style: italic;\">Descriptor identification for material properties via compressed sensing</span>, in preparation.\n</div>\n<p></p>\n \n<p>SIS-SA works iteratively. In the first iteration, a number k of features is collected as those that have the largest correlation (scalar product) with the property vector. The feature with the largest correlation is simply the 1D descriptor. Next, a residual is constructed as the error made at the first iteration. A new set of k features is now selected as those having the largest  correlation with the residual. The 2D descriptor is the pair of features that yield the smallest fitting error upon least square regression, among all possible pairs contained in the union of the sets selected in this and the first iteration. In each next iteration a new residual is constructed as the error made in the previous iteration, then a new set of k features is extracted as those that have largest correlation with each new residual. The nD descriptor is the n-tuple of features that yield the smallest fitting error upon least square regression, among all possible n-tuples contained in the union of the sets obtained in each new iteration and all the previous iterations. If k=1 the method collapses to the so-called orthogonal matching pursuit.\n</p>\n\n        <p>The prediction of the ground-state structure for binary compounds from a simple descriptor has a notable history in materials science [1-7], where descriptors were designed by chemically/physically-inspired intuition. The tool presented here allows for the machine-learning-aided automatic discovery of a descriptor and a model for the prediction of the difference in energy between a selected pair of structures for 82 octet binary materials.</p>\n\n\n        <p> By running the tutorial with the default setting, the (RS vs. ZB) results of the <a href=\"http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.114.105503\" target=\"_blank\">PRL 2015</a> identified by the LASSO+L0 method can be recovered.</p>\n        \n\n        <p>References:</p>\n        <ol>\n          <li>J. A. van Vechten, Phys. Rev. 182, 891 (1969).</li>\n          <li>J. C. Phillips, Rev. Mod. Phys. 42, 317 (1970).</li>\n          <li>J. St. John and A.N. Bloch, Phys. Rev. Lett. 33, 1095 (1974).</li>\n          <li>J. R. Chelikowsky and J. C. Phillips, Phys. Rev. B 17, 2453 (1978).</li>\n          <li>A. Zunger, Phys. Rev. B 22, 5839 (1980).</li>\n          <li>D. G. Pettifor, Solid State Commun. 51, 31 (1984).</li>\n          <li>Y. Saad, D. Gao, T. Ngo, S. Bobbitt, J. R. Chelikowsky, and W. Andreoni, Phys. Rev. B 85, 104104 (2012).</li>\n        </ol>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n<!--         <button type=\"button\" class=\"btn btn-primary\">Save changes</button> -->\n      </div>\n    </div>\n  </div>\n</div>\n\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" data-toggle=\"modal\" data-target=\"#lasso-instructions-modal\">\n Instructions\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<!-- Modal -->\n<div class=\"modal fade\" id=\"lasso-instructions-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"lasso-instructions-modal-label\" style=\"display: none;\">\n  <div class=\"modal-dialog\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"lasso-instructions-modal-label\">Instructions</h4>\n      </div>\n      <div class=\"modal-body lasso_instructions\">\n<p>In this example, you can run a compressed-sensing based algorithm for finding the optimal descriptor and model that predicts the difference in energy between crystal structures (here, rocksalt vs. zincblende vs. CsCl structure). </p>\n\n<p>The descriptor is selected out of a large number of candidates constructed as functions of basic input features, the primary features. </p>\n\n<p>By clicking <b>Settings</b> you can select the structure pair of interest (either RS/ZB or CsCl/ZB), the primary features as well as which kind of unary and binary operations are allowed from the checklist below. Moreover the dimension of the output energies (kcal/mol or eV) and the following three parameters of the SIS+L0 algorithm can be specified: \n        </p><ul>\n          <li>Number of iterations for the construction for the feature space: How often the selected operations are applied to build the feature space. At each step the opreations are applied on all features created untill the current step. </li>\n          <li>Optimal descriptor maximum dimension: Number of SIS+SA iterations.</li>\n          <li>Number of collected features per SIS iteration.</li>\n        </ul>    \n        \n<p></p>\n    \n  \n<p>        After the preferred settings have been adjusted, click <b>RUN</b> for performing the calculations (loading the values of the primary features, creation of the feature space, and optimization via SIS+L0). </p>\n\nDuring the run, a brief summary is printed out below the <b>RUN</b> button. At the end of the run: \n  <ul>\n  <li> the solution (machine-learned descriptor, model, and its performance in terms of training error) is printed out underneath starting from the one-dimensional solution to the selected maximum dimensionality and</li>\n<li> the “View interactive 2D scatter plot” button unlocks; by clicking, the scatter plot with the two-dimensional descriptor appears in a separate tab. In case a dimensionality higher than 2 was selected for the descriptor, the plot displays the two-dimensional descriptor.</li>\n</ul>\n<p>Note: the plot stays active also after another run is performed, so that the output of several sets of input parameters can be compared in the viewer tabs.</p>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n<!--         <button type=\"button\" class=\"btn btn-primary\">Save changes</button> -->\n      </div>\n    </div>\n  </div>\n</div>\n\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" onclick=\"toggle_settings()\">\n Settings\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<a target=\"_blank\" href=\"http://forum.analytics-toolkit.nomad-coe.eu/\" class=\"btn btn-primary\"> Tell us what you think</a>\n\n"
+                    "object": "<script>\nvar beaker = bkHelper.getBeakerObject().beakerObj;\n</script>\n<script>\nvar run_lasso = function() {\n  $(\"#lasso_result_button\").removeClass(\"active\").addClass(\"disabled\");\n  getFeatures();\n  getOperators();\n  beaker.max_dim = $(\"#lasso_max_dim_selector\").val();\n  beaker.structures_diff = $(\"#lasso_structures_diff\").val();\n  beaker.n_comb = $(\"#n_comb\").val();\n  beaker.n_sis = $(\"#n_sis\").val();\n  beaker.units = $(\"#units_select\").val();\n  beaker.evaluate(\"calc_cell\"); // evaluate cells with tag \"lasso_cell\"\n // view_result()\n};\nvar reset_lasso = function(){\n  beaker.evaluate(\"lasso-settings-cell\");\n  var e = document.getElementById('lasso-hidden-settings-div');\n  var b = document.getElementById('lasso-hidden-settings-button');\n  e.style.display = 'block';\n  b.style.display = 'inline';\n};\nvar getFeatures = function() {\n    beaker.selected_feature_list = [];\n    $('#lasso_features_select input:checkbox').each(function () {\n        if(this.checked )\n          beaker.selected_feature_list.push(this.value);\n    });\n};\nvar getOperators = function() {\n    beaker.allowed_operations = [];\n    $('#lasso_operators_select input:checkbox').each(function () {\n        if(this.checked )\n          beaker.allowed_operations.push(this.value);\n    });\n};  \n\nvar toggle_settings = function(){\n  var e = document.getElementById('lasso-hidden-settings-div');\n  var b = document.getElementById('lasso-hidden-settings-button');\n  if(e.style.display == 'block'){\n    e.style.display = 'none';\n    b.style.display = 'none';\n  }\n  else{\n    e.style.display = 'block';\n    b.style.display = 'inline';\n  }\n};\nbeaker.view_result = function(result_link) {\n//   beaker.evaluate(\"lasso_viewer_result\").then(function(x) {\n    $(\"#lasso_result_button\").attr(\"href\", result_link);\n//   }); \n  $(\"#lasso_result_button\").removeClass(\"disabled\").addClass(\"active\");\n}\n\n\n\n\n\n\n\n</script>\n<style type=\"text/css\">\n .lasso_instructions{\n    font-size: 15px;\n  } \n</style>\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" data-toggle=\"modal\" data-target=\"#lasso-motivation-modal\">\n Background\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<!-- Modal -->\n<div class=\"modal fade\" id=\"lasso-motivation-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"lasso-motivation-modal-label\">\n  <div class=\"modal-dialog modal-lg\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"lasso-motivation-modal-label\">Background</h4>\n      </div>\n      <div class=\"modal-body lasso_instructions\">\n        <p>We present a tool for predicting the crystal structure of octet binary compounds, by using a set of descriptive parameters (a descriptor) based on free-atom data of the atomic species constituting the binary material. This task is similar to the one presented in the tutorial <a href=\"http://analytics-toolkit.nomad-coe.eu/tutorial-LASSO_L0\">\"[Predicting energy differences between different crystal structures I: large feature space]\"</a>\". In contrast to that tutorial, here we apply a newly developed method: Sure Independent Screening - Sparse Approximation (SIS-SA), that allows to find an optimal descriptor in a huge feature space containing billions of features.\nThe method is described in:\n</p><div style=\"padding: 1ex; margin-top: 1ex; margin-bottom: 1ex; border-style: dotted; border-width: 1pt; border-color: blue; border-radius: 3px;\">\nR. Ouyang, E. Ahmetcik, L. M. Ghiringhelli, and M. Scheffler: <span style=\"font-style: italic;\">Descriptor identification for material properties via compressed sensing</span>, in preparation.\n</div>\n<p></p>\n \n<p>SIS-SA works iteratively. In the first iteration, a number k of features is collected as those that have the largest correlation (scalar product) with the property vector. The feature with the largest correlation is simply the 1D descriptor. Next, a residual is constructed as the error made at the first iteration. A new set of k features is now selected as those having the largest  correlation with the residual. The 2D descriptor is the pair of features that yield the smallest fitting error upon least square regression, among all possible pairs contained in the union of the sets selected in this and the first iteration. In each next iteration a new residual is constructed as the error made in the previous iteration, then a new set of k features is extracted as those that have largest correlation with each new residual. The nD descriptor is the n-tuple of features that yield the smallest fitting error upon least square regression, among all possible n-tuples contained in the union of the sets obtained in each new iteration and all the previous iterations. If k=1 the method collapses to the so-called orthogonal matching pursuit.\n</p>\n\n        <p>The prediction of the ground-state structure for binary compounds from a simple descriptor has a notable history in materials science [1-7], where descriptors were designed by chemically/physically-inspired intuition. The tool presented here allows for the machine-learning-aided automatic discovery of a descriptor and a model for the prediction of the difference in energy between a selected pair of structures for 82 octet binary materials.</p>\n\n\n        <p> By running the tutorial with the default setting, the (RS vs. ZB) results of the <a href=\"http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.114.105503\" target=\"_blank\">PRL 2015</a> identified by the LASSO+L0 method can be recovered.</p>\n        \n\n        <p>References:</p>\n        <ol>\n          <li>J. A. van Vechten, Phys. Rev. 182, 891 (1969).</li>\n          <li>J. C. Phillips, Rev. Mod. Phys. 42, 317 (1970).</li>\n          <li>J. St. John and A.N. Bloch, Phys. Rev. Lett. 33, 1095 (1974).</li>\n          <li>J. R. Chelikowsky and J. C. Phillips, Phys. Rev. B 17, 2453 (1978).</li>\n          <li>A. Zunger, Phys. Rev. B 22, 5839 (1980).</li>\n          <li>D. G. Pettifor, Solid State Commun. 51, 31 (1984).</li>\n          <li>Y. Saad, D. Gao, T. Ngo, S. Bobbitt, J. R. Chelikowsky, and W. Andreoni, Phys. Rev. B 85, 104104 (2012).</li>\n        </ol>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n<!--         <button type=\"button\" class=\"btn btn-primary\">Save changes</button> -->\n      </div>\n    </div>\n  </div>\n</div>\n\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" data-toggle=\"modal\" data-target=\"#lasso-instructions-modal\">\n Instructions\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<!-- Modal -->\n<div class=\"modal fade\" id=\"lasso-instructions-modal\" tabindex=\"-1\" role=\"dialog\" aria-labelledby=\"lasso-instructions-modal-label\" style=\"display: none;\">\n  <div class=\"modal-dialog\" role=\"document\">\n    <div class=\"modal-content\">\n      <div class=\"modal-header\">\n        <button type=\"button\" class=\"close\" data-dismiss=\"modal\" aria-label=\"Close\"><span aria-hidden=\"true\">×</span></button>\n        <h4 class=\"modal-title\" id=\"lasso-instructions-modal-label\">Instructions</h4>\n      </div>\n      <div class=\"modal-body lasso_instructions\">\n<p>In this example, you can run a compressed-sensing based algorithm for finding the optimal descriptor and model that predicts the difference in energy between crystal structures (here, rocksalt vs. zincblende vs. CsCl structure). </p>\n\n<p>The descriptor is selected out of a large number of candidates constructed as functions of basic input features, the primary features. </p>\n\n<p>By clicking <b>Settings</b> you can select the structure pair of interest (either RS/ZB or CsCl/ZB), the primary features as well as which kind of unary and binary operations are allowed from the checklist below. Moreover the dimension of the output energies (kcal/mol or eV) and the following three parameters of the SIS+L0 algorithm can be specified: \n        </p><ul>\n          <li>Number of iterations for the construction for the feature space: How often the selected operations are applied to build the feature space. At each step the opreations are applied on all features created untill the current step. </li>\n          <li>Optimal descriptor maximum dimension: Number of SIS+SA iterations.</li>\n          <li>Number of collected features per SIS iteration.</li>\n        </ul>    \n        \n<p></p>\n    \n  \n<p>        After the preferred settings have been adjusted, click <b>RUN</b> for performing the calculations (loading the values of the primary features, creation of the feature space, and optimization via SIS+L0). </p>\n\nDuring the run, a brief summary is printed out below the <b>RUN</b> button. At the end of the run: \n  <ul>\n  <li> the solution (machine-learned descriptor, model, and its performance in terms of training error) is printed out underneath starting from the one-dimensional solution to the selected maximum dimensionality and</li>\n<li> the “View interactive 2D scatter plot” button unlocks; by clicking, the scatter plot with the two-dimensional descriptor appears in a separate tab. In case a dimensionality higher than 2 was selected for the descriptor, the plot displays the two-dimensional descriptor.</li>\n</ul>\n<p>Note: the plot stays active also after another run is performed, so that the output of several sets of input parameters can be compared in the viewer tabs.</p>\n      </div>\n      <div class=\"modal-footer\">\n        <button type=\"button\" class=\"btn btn-default\" data-dismiss=\"modal\">Close</button>\n<!--         <button type=\"button\" class=\"btn btn-primary\">Save changes</button> -->\n      </div>\n    </div>\n  </div>\n</div>\n\n<!-- Button trigger modal -->\n<button type=\"button\" class=\"btn btn-default\" onclick=\"toggle_settings()\">\n Settings\n</button>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;\n\n<a target=\"_blank\" href=\"https://nomad-forum.rz-berlin.mpg.de//\" class=\"btn btn-primary\"> Tell us what you think</a>\n\n"
                 },
                 "selectedType": "BeakerDisplay",
                 "elapsedTime": 0,