Mercurial > repos > goeckslab > image_learner
comparison utils.py @ 8:85e6f4b2ad18 draft default tip
planemo upload for repository https://github.com/goeckslab/gleam.git commit 8a42eb9b33df7e1df5ad5153b380e20b910a05b6
author | goeckslab |
---|---|
date | Thu, 14 Aug 2025 14:53:10 +0000 |
parents | 186424a7eca7 |
children |
comparison
equal
deleted
inserted
replaced
7:801a8b6973fb | 8:85e6f4b2ad18 |
---|---|
6 return """ | 6 return """ |
7 <html> | 7 <html> |
8 <head> | 8 <head> |
9 <meta charset="UTF-8"> | 9 <meta charset="UTF-8"> |
10 <title>Galaxy-Ludwig Report</title> | 10 <title>Galaxy-Ludwig Report</title> |
11 | |
12 <!-- your existing styles --> | |
11 <style> | 13 <style> |
12 body { | 14 body { |
13 font-family: Arial, sans-serif; | 15 font-family: Arial, sans-serif; |
14 margin: 0; | 16 margin: 0; |
15 padding: 20px; | 17 padding: 20px; |
30 h2 { | 32 h2 { |
31 border-bottom: 2px solid #4CAF50; | 33 border-bottom: 2px solid #4CAF50; |
32 color: #4CAF50; | 34 color: #4CAF50; |
33 padding-bottom: 5px; | 35 padding-bottom: 5px; |
34 } | 36 } |
37 /* baseline table setup */ | |
35 table { | 38 table { |
36 border-collapse: collapse; | 39 border-collapse: collapse; |
37 margin: 20px 0; | 40 margin: 20px 0; |
38 width: 100%; | 41 width: 100%; |
39 table-layout: fixed; /* Enforces consistent column widths */ | 42 table-layout: fixed; |
40 } | 43 } |
41 table, th, td { | 44 table, th, td { |
42 border: 1px solid #ddd; | 45 border: 1px solid #ddd; |
43 } | 46 } |
44 th, td { | 47 th, td { |
45 padding: 8px; | 48 padding: 8px; |
46 text-align: center; /* Center-align text */ | 49 text-align: center; |
47 vertical-align: middle; /* Center-align content vertically */ | 50 vertical-align: middle; |
48 word-wrap: break-word; /* Break long words to avoid overflow */ | 51 word-wrap: break-word; |
49 } | |
50 th:first-child, td:first-child { | |
51 width: 5%; /* Smaller width for the first column */ | |
52 } | |
53 th:nth-child(2), td:nth-child(2) { | |
54 width: 50%; /* Wider for the metric/description column */ | |
55 } | |
56 th:last-child, td:last-child { | |
57 width: 25%; /* Value column gets remaining space */ | |
58 } | 52 } |
59 th { | 53 th { |
60 background-color: #4CAF50; | 54 background-color: #4CAF50; |
61 color: white; | 55 color: white; |
62 } | 56 } |
66 } | 60 } |
67 .plot img { | 61 .plot img { |
68 max-width: 100%; | 62 max-width: 100%; |
69 height: auto; | 63 height: auto; |
70 } | 64 } |
65 | |
66 /* ------------------- | |
67 SORTABLE COLUMNS | |
68 ------------------- */ | |
69 table.performance-summary th.sortable { | |
70 cursor: pointer; | |
71 position: relative; | |
72 user-select: none; | |
73 } | |
74 /* hide arrows by default */ | |
75 table.performance-summary th.sortable::after { | |
76 content: ''; | |
77 position: absolute; | |
78 right: 12px; | |
79 top: 50%; | |
80 transform: translateY(-50%); | |
81 font-size: 0.8em; | |
82 color: #666; | |
83 } | |
84 /* three states */ | |
85 table.performance-summary th.sortable.sorted-none::after { | |
86 content: '⇅'; | |
87 } | |
88 table.performance-summary th.sortable.sorted-asc::after { | |
89 content: '↑'; | |
90 } | |
91 table.performance-summary th.sortable.sorted-desc::after { | |
92 content: '↓'; | |
93 } | |
71 </style> | 94 </style> |
95 | |
96 <!-- sorting script --> | |
97 <script> | |
98 document.addEventListener('DOMContentLoaded', () => { | |
99 // 1) record each row's original position | |
100 document.querySelectorAll('table.performance-summary tbody').forEach(tbody => { | |
101 Array.from(tbody.rows).forEach((row, i) => { | |
102 row.dataset.originalOrder = i; | |
103 }); | |
104 }); | |
105 | |
106 const getText = cell => cell.innerText.trim(); | |
107 const comparer = (idx, asc) => (a, b) => { | |
108 const v1 = getText(a.children[idx]); | |
109 const v2 = getText(b.children[idx]); | |
110 const n1 = parseFloat(v1), n2 = parseFloat(v2); | |
111 if (!isNaN(n1) && !isNaN(n2)) { | |
112 return asc ? n1 - n2 : n2 - n1; | |
113 } | |
114 return asc | |
115 ? v1.localeCompare(v2) | |
116 : v2.localeCompare(v1); | |
117 }; | |
118 | |
119 document | |
120 .querySelectorAll('table.performance-summary th.sortable') | |
121 .forEach(th => { | |
122 // initialize to "none" state | |
123 th.classList.add('sorted-none'); | |
124 th.addEventListener('click', () => { | |
125 const table = th.closest('table'); | |
126 const allTh = table.querySelectorAll('th.sortable'); | |
127 | |
128 // 1) determine current state BEFORE clearing classes | |
129 let curr = th.classList.contains('sorted-asc') | |
130 ? 'asc' | |
131 : th.classList.contains('sorted-desc') | |
132 ? 'desc' | |
133 : 'none'; | |
134 // 2) cycle to next state | |
135 let next = curr === 'none' | |
136 ? 'asc' | |
137 : curr === 'asc' | |
138 ? 'desc' | |
139 : 'none'; | |
140 | |
141 // 3) clear all sort markers | |
142 allTh.forEach(h => | |
143 h.classList.remove('sorted-none','sorted-asc','sorted-desc') | |
144 ); | |
145 // 4) apply the new marker | |
146 th.classList.add(`sorted-${next}`); | |
147 | |
148 // 5) sort or restore original order | |
149 const tbody = table.querySelector('tbody'); | |
150 let rows = Array.from(tbody.rows); | |
151 if (next === 'none') { | |
152 rows.sort((a, b) => | |
153 a.dataset.originalOrder - b.dataset.originalOrder | |
154 ); | |
155 } else { | |
156 const idx = Array.from(th.parentNode.children).indexOf(th); | |
157 rows.sort(comparer(idx, next === 'asc')); | |
158 } | |
159 rows.forEach(r => tbody.appendChild(r)); | |
160 }); | |
161 }); | |
162 }); | |
163 </script> | |
72 </head> | 164 </head> |
73 <body> | 165 <body> |
74 <div class="container"> | 166 <div class="container"> |
75 """ | 167 """ |
76 | 168 |
201 display: block; | 293 display: block; |
202 }} | 294 }} |
203 </style> | 295 </style> |
204 | 296 |
205 <div class="tabs"> | 297 <div class="tabs"> |
206 <div class="tab active" onclick="showTab('metrics')">Config & Results Summary</div> | 298 <div class="tab active" onclick="showTab('metrics')">Config and Results Summary</div> |
207 <div class="tab" onclick="showTab('trainval')">Train/Validation Results</div> | 299 <div class="tab" onclick="showTab('trainval')">Train/Validation Results</div> |
208 <div class="tab" onclick="showTab('test')">Test Results</div> | 300 <div class="tab" onclick="showTab('test')">Test Results</div> |
209 <!-- always-visible help button --> | 301 <!-- always-visible help button --> |
210 <button id="openMetricsHelp" class="help-btn">Help</button> | 302 <button id="openMetricsHelp" class="help-btn">Help</button> |
211 </div> | 303 </div> |
230 </script> | 322 </script> |
231 """ | 323 """ |
232 | 324 |
233 | 325 |
234 def get_metrics_help_modal() -> str: | 326 def get_metrics_help_modal() -> str: |
235 modal_html = """ | 327 modal_html = ( |
236 <div id="metricsHelpModal" class="modal"> | 328 '<div id="metricsHelpModal" class="modal">' |
237 <div class="modal-content"> | 329 ' <div class="modal-content">' |
238 <span class="close">×</span> | 330 ' <span class="close">×</span>' |
239 <h2>Model Evaluation Metrics — Help Guide</h2> | 331 ' <h2>Model Evaluation Metrics — Help Guide</h2>' |
240 <div class="metrics-guide"> | 332 ' <div class="metrics-guide">' |
241 <h3>1) General Metrics</h3> | 333 ' <h3>1) General Metrics (Regression and Classification)</h3>' |
242 <p><strong>Loss:</strong> Measures the difference between predicted and actual values. Lower is better. Often used for optimization during training.</p> | 334 ' <p><strong>Loss (Regression & Classification):</strong> ' |
243 <p><strong>Accuracy:</strong> Proportion of correct predictions among all predictions. Simple but can be misleading for imbalanced datasets.</p> | 335 'Measures the difference between predicted and actual values, ' |
244 <p><strong>Micro Accuracy:</strong> Calculates accuracy by summing up all individual true positives and true negatives across all classes, making it suitable for multiclass or multilabel problems.</p> | 336 'optimized during training. Lower is better. ' |
245 <p><strong>Token Accuracy:</strong> Measures how often the predicted tokens (e.g., in sequences) match the true tokens. Useful in sequence prediction tasks like NLP.</p> | 337 'For regression, this is often Mean Squared Error (MSE) or ' |
246 <h3>2) Precision, Recall & Specificity</h3> | 338 'Mean Absolute Error (MAE). For classification, it’s typically ' |
247 <p><strong>Precision:</strong> Out of all positive predictions, how many were correct. Precision = TP / (TP + FP). Helps when false positives are costly.</p> | 339 'cross-entropy or log loss.</p>' |
248 <p><strong>Recall (Sensitivity):</strong> Out of all actual positives, how many were predicted correctly. Recall = TP / (TP + FN). Important when missing positives is risky.</p> | 340 ' <h3>2) Regression Metrics</h3>' |
249 <p><strong>Specificity:</strong> True negative rate. Measures how well the model identifies negatives. Specificity = TN / (TN + FP). Useful in medical testing to avoid false alarms.</p> | 341 ' <p><strong>Mean Absolute Error (MAE):</strong> ' |
250 <h3>3) Macro, Micro, and Weighted Averages</h3> | 342 'Average of absolute differences between predicted and actual values, ' |
251 <p><strong>Macro Precision / Recall / F1:</strong> Averages the metric across all classes, treating each class equally, regardless of class frequency. Best when class sizes are balanced.</p> | 343 'in the same units as the target. Use for interpretable error measurement ' |
252 <p><strong>Micro Precision / Recall / F1:</strong> Aggregates TP, FP, FN across all classes before computing the metric. Gives a global view and is ideal for class-imbalanced problems.</p> | 344 'when all errors are equally important. Less sensitive to outliers than MSE.</p>' |
253 <p><strong>Weighted Precision / Recall / F1:</strong> Averages each metric across classes, weighted by the number of true instances per class. Balances importance of classes based on frequency.</p> | 345 ' <p><strong>Mean Squared Error (MSE):</strong> ' |
254 <h3>4) Average Precision (PR-AUC Variants)</h3> | 346 'Average of squared differences between predicted and actual values. ' |
255 <p><strong>Average Precision Macro:</strong> Precision-Recall AUC averaged across all classes equally. Useful for balanced multi-class problems.</p> | 347 'Penalizes larger errors more heavily, useful when large deviations are critical. ' |
256 <p><strong>Average Precision Micro:</strong> Global Precision-Recall AUC using all instances. Best for imbalanced data or multi-label classification.</p> | 348 'Often used as the loss function in regression.</p>' |
257 <p><strong>Average Precision Samples:</strong> Precision-Recall AUC averaged across individual samples (not classes). Ideal for multi-label problems where each sample can belong to multiple classes.</p> | 349 ' <p><strong>Root Mean Squared Error (RMSE):</strong> ' |
258 <h3>5) ROC-AUC Variants</h3> | 350 'Square root of MSE, in the same units as the target. ' |
259 <p><strong>ROC-AUC:</strong> Measures model's ability to distinguish between classes. AUC = 1 is perfect; 0.5 is random guessing. Use for binary classification.</p> | 351 'Balances interpretability and sensitivity to large errors. ' |
260 <p><strong>Macro ROC-AUC:</strong> Averages the AUC across all classes equally. Suitable when classes are balanced and of equal importance.</p> | 352 'Widely used for regression evaluation.</p>' |
261 <p><strong>Micro ROC-AUC:</strong> Computes AUC from aggregated predictions across all classes. Useful in multiclass or multilabel settings with imbalance.</p> | 353 ' <p><strong>Mean Absolute Percentage Error (MAPE):</strong> ' |
262 <h3>6) Ranking Metrics</h3> | 354 'Average absolute error as a percentage of actual values. ' |
263 <p><strong>Hits at K:</strong> Measures whether the true label is among the top-K predictions. Common in recommendation systems and retrieval tasks.</p> | 355 'Scale-independent, ideal for comparing relative errors across datasets. ' |
264 <h3>7) Confusion Matrix Stats (Per Class)</h3> | 356 'Avoid when actual values are near zero.</p>' |
265 <p><strong>True Positives / Negatives (TP / TN):</strong> Correct predictions for positives and negatives respectively.</p> | 357 ' <p><strong>Root Mean Squared Percentage Error (RMSPE):</strong> ' |
266 <p><strong>False Positives / Negatives (FP / FN):</strong> Incorrect predictions — false alarms and missed detections.</p> | 358 'Square root of mean squared percentage error. Scale-independent, ' |
267 <h3>8) Other Useful Metrics</h3> | 359 'penalizes larger relative errors more than MAPE. Use for forecasting ' |
268 <p><strong>Cohen's Kappa:</strong> Measures agreement between predicted and actual values adjusted for chance. Useful for multiclass classification with imbalanced labels.</p> | 360 'or when relative accuracy matters.</p>' |
269 <p><strong>Matthews Correlation Coefficient (MCC):</strong> Balanced measure of prediction quality that takes into account TP, TN, FP, and FN. Particularly effective for imbalanced datasets.</p> | 361 ' <p><strong>R² Score:</strong> Proportion of variance in the target ' |
270 <h3>9) Metric Recommendations</h3> | 362 'explained by the model. Ranges from negative infinity to 1 (perfect prediction). ' |
271 <ul> | 363 'Use to assess model fit; negative values indicate poor performance ' |
272 <li>Use <strong>Accuracy + F1</strong> for balanced data.</li> | 364 'compared to predicting the mean.</p>' |
273 <li>Use <strong>Precision, Recall, ROC-AUC</strong> for imbalanced datasets.</li> | 365 ' <h3>3) Classification Metrics</h3>' |
274 <li>Use <strong>Average Precision Micro</strong> for multilabel or class-imbalanced problems.</li> | 366 ' <p><strong>Accuracy:</strong> Proportion of correct predictions ' |
275 <li>Use <strong>Macro scores</strong> when all classes should be treated equally.</li> | 367 'among all predictions. Simple but misleading for imbalanced datasets, ' |
276 <li>Use <strong>Weighted scores</strong> when class imbalance should be accounted for without ignoring small classes.</li> | 368 'where high accuracy may hide poor performance on minority classes.</p>' |
277 <li>Use <strong>Confusion Matrix stats</strong> to analyze class-wise performance.</li> | 369 ' <p><strong>Micro Accuracy:</strong> Sums true positives and true negatives ' |
278 <li>Use <strong>Hits at K</strong> for recommendation or ranking-based tasks.</li> | 370 'across all classes before computing accuracy. Suitable for multiclass or ' |
279 </ul> | 371 'multilabel problems with imbalanced data.</p>' |
280 </div> | 372 ' <p><strong>Token Accuracy:</strong> Measures how often predicted tokens ' |
281 </div> | 373 '(e.g., in sequences) match true tokens. Common in NLP tasks like text generation ' |
282 </div> | 374 'or token classification.</p>' |
283 """ | 375 ' <p><strong>Precision:</strong> Proportion of positive predictions that are ' |
284 modal_css = """ | 376 'correct (TP / (TP + FP)). Use when false positives are costly, e.g., spam detection.</p>' |
285 <style> | 377 ' <p><strong>Recall (Sensitivity):</strong> Proportion of actual positives ' |
286 .modal { | 378 'correctly predicted (TP / (TP + FN)). Use when missing positives is risky, ' |
287 display: none; | 379 'e.g., disease detection.</p>' |
288 position: fixed; | 380 ' <p><strong>Specificity:</strong> True negative rate (TN / (TN + FP)). ' |
289 z-index: 1; | 381 'Measures ability to identify negatives. Useful in medical testing to avoid ' |
290 left: 0; | 382 'false alarms.</p>' |
291 top: 0; | 383 ' <h3>4) Classification: Macro, Micro, and Weighted Averages</h3>' |
292 width: 100%; | 384 ' <p><strong>Macro Precision / Recall / F1:</strong> Averages the metric ' |
293 height: 100%; | 385 'across all classes, treating each equally. Best for balanced datasets where ' |
294 overflow: auto; | 386 'all classes are equally important.</p>' |
295 background-color: rgba(0,0,0,0.4); | 387 ' <p><strong>Micro Precision / Recall / F1:</strong> Aggregates true positives, ' |
296 } | 388 'false positives, and false negatives across all classes before computing. ' |
297 .modal-content { | 389 'Ideal for imbalanced or multilabel classification.</p>' |
298 background-color: #fefefe; | 390 ' <p><strong>Weighted Precision / Recall / F1:</strong> Averages metrics ' |
299 margin: 15% auto; | 391 'across classes, weighted by the number of true instances per class. Balances ' |
300 padding: 20px; | 392 'class importance based on frequency.</p>' |
301 border: 1px solid #888; | 393 ' <h3>5) Classification: Average Precision (PR-AUC Variants)</h3>' |
302 width: 80%; | 394 ' <p><strong>Average Precision Macro:</strong> Precision-Recall AUC averaged ' |
303 max-width: 800px; | 395 'equally across classes. Use for balanced multiclass problems.</p>' |
304 } | 396 ' <p><strong>Average Precision Micro:</strong> Global Precision-Recall AUC ' |
305 .close { | 397 'using all instances. Best for imbalanced or multilabel classification.</p>' |
306 color: #aaa; | 398 ' <p><strong>Average Precision Samples:</strong> Precision-Recall AUC averaged ' |
307 float: right; | 399 'across individual samples. Ideal for multilabel tasks where samples have multiple ' |
308 font-size: 28px; | 400 'labels.</p>' |
309 font-weight: bold; | 401 ' <h3>6) Classification: ROC-AUC Variants</h3>' |
310 } | 402 ' <p><strong>ROC-AUC:</strong> Measures ability to distinguish between classes. ' |
311 .close:hover, | 403 'AUC = 1 is perfect; 0.5 is random guessing. Use for binary classification.</p>' |
312 .close:focus { | 404 ' <p><strong>Macro ROC-AUC:</strong> Averages AUC across all classes equally. ' |
313 color: black; | 405 'Suitable for balanced multiclass problems.</p>' |
314 text-decoration: none; | 406 ' <p><strong>Micro ROC-AUC:</strong> Computes AUC from aggregated predictions ' |
315 cursor: pointer; | 407 'across all classes. Useful for imbalanced or multilabel settings.</p>' |
316 } | 408 ' <h3>7) Classification: Confusion Matrix Stats (Per Class)</h3>' |
317 .metrics-guide h3 { | 409 ' <p><strong>True Positives / Negatives (TP / TN):</strong> Correct predictions ' |
318 margin-top: 20px; | 410 'for positives and negatives, respectively.</p>' |
319 } | 411 ' <p><strong>False Positives / Negatives (FP / FN):</strong> Incorrect predictions ' |
320 .metrics-guide p { | 412 '— false alarms and missed detections.</p>' |
321 margin: 5px 0; | 413 ' <h3>8) Classification: Ranking Metrics</h3>' |
322 } | 414 ' <p><strong>Hits at K:</strong> Measures whether the true label is among the ' |
323 .metrics-guide ul { | 415 'top-K predictions. Common in recommendation systems and retrieval tasks.</p>' |
324 margin: 10px 0; | 416 ' <h3>9) Other Metrics (Classification)</h3>' |
325 padding-left: 20px; | 417 ' <p><strong>Cohen\'s Kappa:</strong> Measures agreement between predicted and ' |
326 } | 418 'actual labels, adjusted for chance. Useful for multiclass classification with ' |
327 </style> | 419 'imbalanced data.</p>' |
328 """ | 420 ' <p><strong>Matthews Correlation Coefficient (MCC):</strong> Balanced measure ' |
329 modal_js = """ | 421 'using TP, TN, FP, and FN. Effective for imbalanced datasets.</p>' |
330 <script> | 422 ' <h3>10) Metric Recommendations</h3>' |
331 document.addEventListener("DOMContentLoaded", function() { | 423 ' <ul>' |
332 var modal = document.getElementById("metricsHelpModal"); | 424 ' <li><strong>Regression:</strong> Use <strong>RMSE</strong> or ' |
333 var openBtn = document.getElementById("openMetricsHelp"); | 425 '<strong>MAE</strong> for general evaluation, <strong>MAPE</strong> for relative ' |
334 var span = document.getElementsByClassName("close")[0]; | 426 'errors, and <strong>R²</strong> to assess model fit. Use <strong>MSE</strong> or ' |
335 if (openBtn && modal) { | 427 '<strong>RMSPE</strong> when large errors are critical.</li>' |
336 openBtn.onclick = function() { | 428 ' <li><strong>Classification (Balanced Data):</strong> Use <strong>Accuracy</strong> ' |
337 modal.style.display = "block"; | 429 'and <strong>F1</strong> for overall performance.</li>' |
338 }; | 430 ' <li><strong>Classification (Imbalanced Data):</strong> Use <strong>Precision</strong>, ' |
339 } | 431 '<strong>Recall</strong>, and <strong>ROC-AUC</strong> to focus on minority class ' |
340 if (span && modal) { | 432 'performance.</li>' |
341 span.onclick = function() { | 433 ' <li><strong>Multilabel or Imbalanced Classification:</strong> Use ' |
342 modal.style.display = "none"; | 434 '<strong>Micro Precision/Recall/F1</strong> or <strong>Micro ROC-AUC</strong>.</li>' |
343 }; | 435 ' <li><strong>Balanced Multiclass:</strong> Use <strong>Macro Precision/Recall/F1</strong> ' |
344 } | 436 'or <strong>Macro ROC-AUC</strong>.</li>' |
345 window.onclick = function(event) { | 437 ' <li><strong>Class Frequency Matters:</strong> Use <strong>Weighted Precision/Recall/F1</strong> ' |
346 if (event.target == modal) { | 438 'to account for class imbalance.</li>' |
347 modal.style.display = "none"; | 439 ' <li><strong>Recommendation/Ranking:</strong> Use <strong>Hits at K</strong> for retrieval tasks.</li>' |
348 } | 440 ' <li><strong>Detailed Analysis:</strong> Use <strong>Confusion Matrix stats</strong> ' |
349 } | 441 'for class-wise performance in classification.</li>' |
350 }); | 442 ' </ul>' |
351 </script> | 443 ' </div>' |
352 """ | 444 ' </div>' |
445 '</div>' | |
446 ) | |
447 modal_css = ( | |
448 "<style>" | |
449 ".modal {" | |
450 " display: none;" | |
451 " position: fixed;" | |
452 " z-index: 1;" | |
453 " left: 0;" | |
454 " top: 0;" | |
455 " width: 100%;" | |
456 " height: 100%;" | |
457 " overflow: auto;" | |
458 " background-color: rgba(0,0,0,0.4);" | |
459 "}" | |
460 ".modal-content {" | |
461 " background-color: #fefefe;" | |
462 " margin: 15% auto;" | |
463 " padding: 20px;" | |
464 " border: 1px solid #888;" | |
465 " width: 80%;" | |
466 " max-width: 800px;" | |
467 "}" | |
468 ".close {" | |
469 " color: #aaa;" | |
470 " float: right;" | |
471 " font-size: 28px;" | |
472 " font-weight: bold;" | |
473 "}" | |
474 ".close:hover," | |
475 ".close:focus {" | |
476 " color: black;" | |
477 " text-decoration: none;" | |
478 " cursor: pointer;" | |
479 "}" | |
480 ".metrics-guide h3 {" | |
481 " margin-top: 20px;" | |
482 "}" | |
483 ".metrics-guide p {" | |
484 " margin: 5px 0;" | |
485 "}" | |
486 ".metrics-guide ul {" | |
487 " margin: 10px 0;" | |
488 " padding-left: 20px;" | |
489 "}" | |
490 "</style>" | |
491 ) | |
492 modal_js = ( | |
493 "<script>" | |
494 'document.addEventListener("DOMContentLoaded", function() {' | |
495 ' var modal = document.getElementById("metricsHelpModal");' | |
496 ' var openBtn = document.getElementById("openMetricsHelp");' | |
497 ' var span = document.getElementsByClassName("close")[0];' | |
498 " if (openBtn && modal) {" | |
499 " openBtn.onclick = function() {" | |
500 " modal.style.display = \"block\";" | |
501 " };" | |
502 " }" | |
503 " if (span && modal) {" | |
504 " span.onclick = function() {" | |
505 " modal.style.display = \"none\";" | |
506 " };" | |
507 " }" | |
508 " window.onclick = function(event) {" | |
509 " if (event.target == modal) {" | |
510 " modal.style.display = \"none\";" | |
511 " }" | |
512 " }" | |
513 "});" | |
514 "</script>" | |
515 ) | |
353 return modal_css + modal_html + modal_js | 516 return modal_css + modal_html + modal_js |