{
  "id": 341813,
  "title": "Competition summary - Congrats to the Winners!",
  "url": "/competitions/energy-anomaly-detection/discussion/341813",
  "author_name": "",
  "post_date": "2022-08-04T11:31:00.858123900Z",
  "votes": 3,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hello Kagglers,</p>\n<p>The Large-scale Energy Anomaly Detection (LEAD) competition has closed! We hope you all liked this exciting problem and were able to learn new machine learning skills!</p>\n<p>In total, we had 75 competitors in 62 teams with more than 600 submissions across the globe! We thank you for your hard work and time commitment to this competition, and congratulations to all winners! We're looking forward to hosting more such competitions with prize money in the future, and we hope you all will compete again.</p>\n<p>We've created this thread to be a curated list of winning solutions summary, and top notebooks and posts. <strong>We highly encourage you to post your solution summary and also share your notebooks for others to benefit</strong>. We will update this post and keep track of all relevant works. </p>\n<p>Additionally, if you write a research article or blog post based on this competition and dataset, and would like to share them with the Kaggle community, let us know, and we will post them here. Finally, if you are interested in collaborating with us to extend this competition dataset or experiment with more advanced energy anomaly detection models, and for possible joint writing of research articles, please contact us.</p>\n<p>Many thanks!<br>\n<a href=\"https://github.com/samy101/lead-dataset\" target=\"_blank\">The LEAD project team</a></p>\n<p><strong>Top solutions</strong></p>\n<ul>\n<li><a href=\"https://www.kaggle.com/competitions/energy-anomaly-detection/discussion/341103\" target=\"_blank\">1st place solution</a></li>\n</ul>\n<p><strong>Top notebooks</strong></p>\n<table>\n<thead>\n<tr>\n<th>S.No</th>\n<th>Notebook</th>\n<th>Public Score</th>\n<th>Private Score</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>1.</td>\n<td><a href=\"https://www.kaggle.com/code/patrick0302/2-anomaly-detection-features-csv-revised-xgb\" target=\"_blank\">XGBoost with lag features</a></td>\n<td>0.95701</td>\n<td>0.97890</td>\n</tr>\n<tr>\n<td>2.</td>\n<td><a href=\"https://www.kaggle.com/code/patrick0302/2-anomaly-detection-features-csv-revised-lgb\" target=\"_blank\">LightGBM with lag features</a></td>\n<td>0.94427</td>\n<td>0.97554</td>\n</tr>\n<tr>\n<td>3.</td>\n<td><a href=\"https://www.kaggle.com/code/patrick0302/2-anomaly-detection-features-csv-revised-ensem\" target=\"_blank\">XGBoost + LightGBM with lag features</a></td>\n<td>0.95698</td>\n<td>0.97987</td>\n</tr>\n<tr>\n<td>4.</td>\n<td><a href=\"https://www.kaggle.com/code/abhishek123maurya/2-anomaly-detection-features-csv\" target=\"_blank\">Logistic Reg. and XGBoost with balanced random samples 1</a></td>\n<td>0.879393</td>\n<td></td>\n</tr>\n<tr>\n<td>5.</td>\n<td><a href=\"https://www.kaggle.com/code/abhishek123maurya/1-anomaly-detection-train-csv\" target=\"_blank\">Logistic Reg. and XGBoost with balanced random samples 2</a></td>\n<td>0.79714</td>\n<td>0.86700</td>\n</tr>\n<tr>\n<td>6.</td>\n<td><a href=\"https://www.kaggle.com/code/abdallahmohamedamin/large-scale-energy-anomaly-detection-lead\" target=\"_blank\">Isolation Forest, KNN, DT, Extra Trees, Random Forest, etc.</a></td>\n<td>0.76331</td>\n<td>0.81887</td>\n</tr>\n<tr>\n<td>7.</td>\n<td><a href=\"https://www.kaggle.com/code/fdalforno/baseline-eda-analisys\" target=\"_blank\">Random Forest and baseline EDA analysis</a></td>\n<td>0.72747</td>\n<td>0.75656</td>\n</tr>\n<tr>\n<td>8.</td>\n<td><a href=\"https://www.kaggle.com/code/rohannanaware/energy-anomaly-detection\" target=\"_blank\">XGBoost and EDA (using R)</a></td>\n<td>0.71054</td>\n<td></td>\n</tr>\n<tr>\n<td>9.</td>\n<td><a href=\"https://www.kaggle.com/code/bigdataanalysis1/anomalize-with-r-and-predict-with-xgboost-one\" target=\"_blank\">XGBoost on H2O (using R)</a></td>\n<td>0.60215</td>\n<td></td>\n</tr>\n<tr>\n<td>10.</td>\n<td><a href=\"https://www.kaggle.com/code/patrick0302/lead-abnormal-viz\" target=\"_blank\">Anomaly Visualization </a></td>\n<td></td>\n<td></td>\n</tr>\n</tbody>\n</table>\n<p><strong>Top discussions</strong> </p>\n<ul>\n<li><a href=\"https://www.kaggle.com/competitions/energy-anomaly-detection/discussion/331301\" target=\"_blank\">Metric on Judging AUC/Score</a></li>\n<li><a href=\"https://www.kaggle.com/competitions/energy-anomaly-detection/discussion/335735\" target=\"_blank\">Building IDs Unique between Train &amp; Test - Implications</a></li>\n</ul>",
  "messages": [
    {
      "id": "1884315",
      "postDate": "08/04/2022 11:31:00",
      "content": "<p>Hello Kagglers,</p>\n<p>The Large-scale Energy Anomaly Detection (LEAD) competition has closed! We hope you all liked this exciting problem and were able to learn new machine learning skills!</p>\n<p>In total, we had 75 competitors in 62 teams with more than 600 submissions across the globe! We thank you for your hard work and time commitment to this competition, and congratulations to all winners! We're looking forward to hosting more such competitions with prize money in the future, and we hope you all will compete again.</p>\n<p>We've created this thread to be a curated list of winning solutions summary, and top notebooks and posts. <strong>We highly encourage you to post your solution summary and also share your notebooks for others to benefit</strong>. We will update this post and keep track of all relevant works. </p>\n<p>Additionally, if you write a research article or blog post based on this competition and dataset, and would like to share them with the Kaggle community, let us know, and we will post them here. Finally, if you are interested in collaborating with us to extend this competition dataset or experiment with more advanced energy anomaly detection models, and for possible joint writing of research articles, please contact us.</p>\n<p>Many thanks!<br>\n<a href=\"https://github.com/samy101/lead-dataset\" target=\"_blank\">The LEAD project team</a></p>\n<p><strong>Top solutions</strong></p>\n<ul>\n<li><a href=\"https://www.kaggle.com/competitions/energy-anomaly-detection/discussion/341103\" target=\"_blank\">1st place solution</a></li>\n</ul>\n<p><strong>Top notebooks</strong></p>\n<table>\n<thead>\n<tr>\n<th>S.No</th>\n<th>Notebook</th>\n<th>Public Score</th>\n<th>Private Score</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>1.</td>\n<td><a href=\"https://www.kaggle.com/code/patrick0302/2-anomaly-detection-features-csv-revised-xgb\" target=\"_blank\">XGBoost with lag features</a></td>\n<td>0.95701</td>\n<td>0.97890</td>\n</tr>\n<tr>\n<td>2.</td>\n<td><a href=\"https://www.kaggle.com/code/patrick0302/2-anomaly-detection-features-csv-revised-lgb\" target=\"_blank\">LightGBM with lag features</a></td>\n<td>0.94427</td>\n<td>0.97554</td>\n</tr>\n<tr>\n<td>3.</td>\n<td><a href=\"https://www.kaggle.com/code/patrick0302/2-anomaly-detection-features-csv-revised-ensem\" target=\"_blank\">XGBoost + LightGBM with lag features</a></td>\n<td>0.95698</td>\n<td>0.97987</td>\n</tr>\n<tr>\n<td>4.</td>\n<td><a href=\"https://www.kaggle.com/code/abhishek123maurya/2-anomaly-detection-features-csv\" target=\"_blank\">Logistic Reg. and XGBoost with balanced random samples 1</a></td>\n<td>0.879393</td>\n<td></td>\n</tr>\n<tr>\n<td>5.</td>\n<td><a href=\"https://www.kaggle.com/code/abhishek123maurya/1-anomaly-detection-train-csv\" target=\"_blank\">Logistic Reg. and XGBoost with balanced random samples 2</a></td>\n<td>0.79714</td>\n<td>0.86700</td>\n</tr>\n<tr>\n<td>6.</td>\n<td><a href=\"https://www.kaggle.com/code/abdallahmohamedamin/large-scale-energy-anomaly-detection-lead\" target=\"_blank\">Isolation Forest, KNN, DT, Extra Trees, Random Forest, etc.</a></td>\n<td>0.76331</td>\n<td>0.81887</td>\n</tr>\n<tr>\n<td>7.</td>\n<td><a href=\"https://www.kaggle.com/code/fdalforno/baseline-eda-analisys\" target=\"_blank\">Random Forest and baseline EDA analysis</a></td>\n<td>0.72747</td>\n<td>0.75656</td>\n</tr>\n<tr>\n<td>8.</td>\n<td><a href=\"https://www.kaggle.com/code/rohannanaware/energy-anomaly-detection\" target=\"_blank\">XGBoost and EDA (using R)</a></td>\n<td>0.71054</td>\n<td></td>\n</tr>\n<tr>\n<td>9.</td>\n<td><a href=\"https://www.kaggle.com/code/bigdataanalysis1/anomalize-with-r-and-predict-with-xgboost-one\" target=\"_blank\">XGBoost on H2O (using R)</a></td>\n<td>0.60215</td>\n<td></td>\n</tr>\n<tr>\n<td>10.</td>\n<td><a href=\"https://www.kaggle.com/code/patrick0302/lead-abnormal-viz\" target=\"_blank\">Anomaly Visualization </a></td>\n<td></td>\n<td></td>\n</tr>\n</tbody>\n</table>\n<p><strong>Top discussions</strong> </p>\n<ul>\n<li><a href=\"https://www.kaggle.com/competitions/energy-anomaly-detection/discussion/331301\" target=\"_blank\">Metric on Judging AUC/Score</a></li>\n<li><a href=\"https://www.kaggle.com/competitions/energy-anomaly-detection/discussion/335735\" target=\"_blank\">Building IDs Unique between Train &amp; Test - Implications</a></li>\n</ul>",
      "rawMarkdown": "Hello Kagglers,\n \nThe Large-scale Energy Anomaly Detection (LEAD) competition has closed! We hope you all liked this exciting problem and were able to learn new machine learning skills!\n \nIn total, we had 75 competitors in 62 teams with more than 600 submissions across the globe! We thank you for your hard work and time commitment to this competition, and congratulations to all winners! We're looking forward to hosting more such competitions with prize money in the future, and we hope you all will compete again.\n \nWe've created this thread to be a curated list of winning solutions summary, and top notebooks and posts. **We highly encourage you to post your solution summary and also share your notebooks for others to benefit**. We will update this post and keep track of all relevant works. \n \nAdditionally, if you write a research article or blog post based on this competition and dataset, and would like to share them with the Kaggle community, let us know, and we will post them here. Finally, if you are interested in collaborating with us to extend this competition dataset or experiment with more advanced energy anomaly detection models, and for possible joint writing of research articles, please contact us.\n\nMany thanks!\n[The LEAD project team](https://github.com/samy101/lead-dataset)\n\n**Top solutions**\n- [1st place solution](https://www.kaggle.com/competitions/energy-anomaly-detection/discussion/341103)\n\n\n**Top notebooks**\n\n| S.No | Notebook | Public Score | Private Score |\n| --- | --- | --- | --- |\n| 1. | [XGBoost with lag features](https://www.kaggle.com/code/patrick0302/2-anomaly-detection-features-csv-revised-xgb)  | 0.95701 | 0.97890 |\n| 2. | [LightGBM with lag features](https://www.kaggle.com/code/patrick0302/2-anomaly-detection-features-csv-revised-lgb) | 0.94427 | 0.97554 |\n| 3. | [XGBoost + LightGBM with lag features](https://www.kaggle.com/code/patrick0302/2-anomaly-detection-features-csv-revised-ensem) | 0.95698 | 0.97987 |\n| 4. | [Logistic Reg. and XGBoost with balanced random samples 1](https://www.kaggle.com/code/abhishek123maurya/2-anomaly-detection-features-csv) | 0.879393 | |\n| 5. | [Logistic Reg. and XGBoost with balanced random samples 2](https://www.kaggle.com/code/abhishek123maurya/1-anomaly-detection-train-csv) | 0.79714 | 0.86700|\n| 6. | [Isolation Forest, KNN, DT, Extra Trees, Random Forest, etc.](https://www.kaggle.com/code/abdallahmohamedamin/large-scale-energy-anomaly-detection-lead)| 0.76331 | 0.81887 |\n| 7. | [Random Forest and baseline EDA analysis](https://www.kaggle.com/code/fdalforno/baseline-eda-analisys) | 0.72747 | 0.75656 |\n|8. | [XGBoost and EDA (using R)](https://www.kaggle.com/code/rohannanaware/energy-anomaly-detection) | 0.71054 | |\n|9. | [XGBoost on H2O (using R)](https://www.kaggle.com/code/bigdataanalysis1/anomalize-with-r-and-predict-with-xgboost-one) | 0.60215 | |\n|10. | [Anomaly Visualization ](https://www.kaggle.com/code/patrick0302/lead-abnormal-viz) | | |\n\n**Top discussions** \n - [Metric on Judging AUC/Score](https://www.kaggle.com/competitions/energy-anomaly-detection/discussion/331301)\n - [Building IDs Unique between Train & Test - Implications](https://www.kaggle.com/competitions/energy-anomaly-detection/discussion/335735)",
      "votes": null
    },
    {
      "id": "1949376",
      "postDate": "09/21/2022 16:31:47",
      "content": "<p><a href=\"https://www.kaggle.com/samy101\" target=\"_blank\">@samy101</a> </p>\n<p>Hi Samy,</p>\n<p>Can you add tags to this competition?  It will help future users find the competition and research its solutions.  eg:   <a href=\"https://www.kaggle.com/code/kaggleqrdl/explorer-widget?scriptVersionId=106216518\" target=\"_blank\">https://www.kaggle.com/code/kaggleqrdl/explorer-widget?scriptVersionId=106216518</a></p>",
      "rawMarkdown": "samy101 \n\nHi Samy,\n\nCan you add tags to this competition?  It will help future users find the competition and research its solutions.  eg:   https://www.kaggle.com/code/kaggleqrdl/explorer-widget?scriptVersionId=106216518",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1949376,
      "author_name": "kaggleqrdl",
      "author_url": "",
      "post_date": "09/21/2022 16:31:47",
      "content": "<p><a href=\"https://www.kaggle.com/samy101\" target=\"_blank\">@samy101</a> </p>\n<p>Hi Samy,</p>\n<p>Can you add tags to this competition?  It will help future users find the competition and research its solutions.  eg:   <a href=\"https://www.kaggle.com/code/kaggleqrdl/explorer-widget?scriptVersionId=106216518\" target=\"_blank\">https://www.kaggle.com/code/kaggleqrdl/explorer-widget?scriptVersionId=106216518</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1884315": "Hello Kagglers,\n \nThe Large-scale Energy Anomaly Detection (LEAD) competition has closed! We hope you all liked this exciting problem and were able to learn new machine learning skills!\n \nIn total, we had 75 competitors in 62 teams with more than 600 submissions across the globe! We thank you for your hard work and time commitment to this competition, and congratulations to all winners! We're looking forward to hosting more such competitions with prize money in the future, and we hope you all will compete again.\n \nWe've created this thread to be a curated list of winning solutions summary, and top notebooks and posts. **We highly encourage you to post your solution summary and also share your notebooks for others to benefit**. We will update this post and keep track of all relevant works. \n \nAdditionally, if you write a research article or blog post based on this competition and dataset, and would like to share them with the Kaggle community, let us know, and we will post them here. Finally, if you are interested in collaborating with us to extend this competition dataset or experiment with more advanced energy anomaly detection models, and for possible joint writing of research articles, please contact us.\n\nMany thanks!\n[The LEAD project team](https://github.com/samy101/lead-dataset)\n\n**Top solutions**\n- [1st place solution](https://www.kaggle.com/competitions/energy-anomaly-detection/discussion/341103)\n\n\n**Top notebooks**\n\n| S.No | Notebook | Public Score | Private Score |\n| --- | --- | --- | --- |\n| 1. | [XGBoost with lag features](https://www.kaggle.com/code/patrick0302/2-anomaly-detection-features-csv-revised-xgb)  | 0.95701 | 0.97890 |\n| 2. | [LightGBM with lag features](https://www.kaggle.com/code/patrick0302/2-anomaly-detection-features-csv-revised-lgb) | 0.94427 | 0.97554 |\n| 3. | [XGBoost + LightGBM with lag features](https://www.kaggle.com/code/patrick0302/2-anomaly-detection-features-csv-revised-ensem) | 0.95698 | 0.97987 |\n| 4. | [Logistic Reg. and XGBoost with balanced random samples 1](https://www.kaggle.com/code/abhishek123maurya/2-anomaly-detection-features-csv) | 0.879393 | |\n| 5. | [Logistic Reg. and XGBoost with balanced random samples 2](https://www.kaggle.com/code/abhishek123maurya/1-anomaly-detection-train-csv) | 0.79714 | 0.86700|\n| 6. | [Isolation Forest, KNN, DT, Extra Trees, Random Forest, etc.](https://www.kaggle.com/code/abdallahmohamedamin/large-scale-energy-anomaly-detection-lead)| 0.76331 | 0.81887 |\n| 7. | [Random Forest and baseline EDA analysis](https://www.kaggle.com/code/fdalforno/baseline-eda-analisys) | 0.72747 | 0.75656 |\n|8. | [XGBoost and EDA (using R)](https://www.kaggle.com/code/rohannanaware/energy-anomaly-detection) | 0.71054 | |\n|9. | [XGBoost on H2O (using R)](https://www.kaggle.com/code/bigdataanalysis1/anomalize-with-r-and-predict-with-xgboost-one) | 0.60215 | |\n|10. | [Anomaly Visualization ](https://www.kaggle.com/code/patrick0302/lead-abnormal-viz) | | |\n\n**Top discussions** \n - [Metric on Judging AUC/Score](https://www.kaggle.com/competitions/energy-anomaly-detection/discussion/331301)\n - [Building IDs Unique between Train & Test - Implications](https://www.kaggle.com/competitions/energy-anomaly-detection/discussion/335735)",
    "1949376": "samy101 \n\nHi Samy,\n\nCan you add tags to this competition?  It will help future users find the competition and research its solutions.  eg:   https://www.kaggle.com/code/kaggleqrdl/explorer-widget?scriptVersionId=106216518"
  },
  "source": "meta"
}