AIQuest_demand_CatBoost.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"name":"AIQuest_demand_CatBoost.ipynb","provenance":[],"collapsed_sections":[],"authorship_tag":"ABX9TyMZHkwPgbK9fmeaUPhL2nUk"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"},"accelerator":"GPU","widgets":{"application/vnd.jupyter.widget-state+json":{"f50370b6d5624e099320a5cba46ebc1f":{"model_module":"catboost-widget","model_name":"CatboostWidgetModel","model_module_version":"^1.0.0","state":{"_view_name":"CatboostWidgetView","_dom_classes":[],"_model_name":"CatboostWidgetModel","data":{"catboost_info":{"content":{"data":{"meta":{"name":"experiment","parameters":"","test_metrics":[{"best_value":"Min","name":"RMSE"}],"learn_sets":["learn"],"launch_mode":"Train","learn_metrics":[{"best_value":"Min","name":"RMSE"}],"iteration_count":1000,"test_sets":["test"]},"iterations":[{"test":[2.307362869],"passed_time":1.188237518,"iteration":0,"remaining_time":1187.049281,"learn":[3.126034168]},{"test":[2.250489473],"passed_time":2.340657364,"iteration":1,"remaining_time":1167.988025,"learn":[3.083070074]},{"test":[2.215149169],"passed_time":3.506483073,"iteration":2,"remaining_time":1165.321208,"learn":[3.052221785]},{"test":[2.186606944],"passed_time":4.471700589,"iteration":3,"remaining_time":1113.453447,"learn":[3.029682574]},{"test":[2.170743861],"passed_time":5.528324442,"iteration":4,"remaining_time":1100.136564,"learn":[3.013646874]},{"test":[2.159021563],"passed_time":6.589510761,"iteration":5,"remaining_time":1091.662283,"learn":[3.00130718]},{"test":[2.152176644],"passed_time":7.660128095,"iteration":6,"remaining_time":1086.643885,"learn":[2.992113536]},{"test":[2.145822982],"passed_time":8.707600557,"iteration":7,"remaining_time":1079.742469,"learn":[2.984519948]},{"test":[2.142647171],"passed_time":9.958467184,"iteration":8,"remaining_time":1096.537887,"learn":[2.978146856]},{"test":[2.140338902],"passed_time":11.07068683,"iteration":9,"remaining_time":1095.997996,"learn":[2.971651946]},{"test":[2.1373448],"passed_time":12.33022892,"iteration":10,"remaining_time":1108.599673,"learn":[2.967918876]},{"test":[2.126546541],"passed_time":13.62176327,"iteration":11,"remaining_time":1121.525176,"learn":[2.959318317]},{"test":[2.121732833],"passed_time":14.87772883,"iteration":12,"remaining_time":1129.56295,"learn":[2.953226822]},{"test":[2.119501169],"passed_time":15.90717494,"iteration":13,"remaining_time":1120.319606,"learn":[2.949010538]},{"test":[2.119012894],"passed_time":16.89699918,"iteration":14,"remaining_time":1109.569613,"learn":[2.945294843]},{"test":[2.118977916],"passed_time":17.99943315,"iteration":15,"remaining_time":1106.965139,"learn":[2.941420536]},{"test":[2.119441523],"passed_time":19.10450835,"iteration":16,"remaining_time":1104.690101,"learn":[2.939310862]},{"test":[2.118444614],"passed_time":20.2559885,"iteration":17,"remaining_time":1105.076706,"learn":[2.936932579]},{"test":[2.115252378],"passed_time":21.55316907,"iteration":18,"remaining_time":1112.824151,"learn":[2.931834131]},{"test":[2.115248352],"passed_time":22.81771686,"iteration":19,"remaining_time":1118.068126,"learn":[2.927974218]},{"test":[2.116930258],"passed_time":24.00551989,"iteration":20,"remaining_time":1119.114475,"learn":[2.925386721]},{"test":[2.115491845],"passed_time":25.34527029,"iteration":21,"remaining_time":1126.71247,"learn":[2.923369522]},{"test":[2.117038108],"passed_time":26.62082025,"iteration":22,"remaining_time":1130.806147,"learn":[2.918978697]},{"test":[2.11673772],"passed_time":27.845745,"iteration":23,"remaining_time":1132.39363,"learn":[2.914954698]},{"test":[2.115780701],"passed_time":29.01533966,"iteration":24,"remaining_time":1131.598247,"learn":[2.913656646]},{"test":[2.115634649],"passed_time":30.04215725,"iteration":25,"remaining_time":1125.425429,"learn":[2.911492578]},{"test":[2.113314336],"passed_time":31.27300273,"iteration":26,"remaining_time":1126.986358,"learn":[2.909806128]},{"test":[2.113870876],"passed_time":32.36494213,"iteration":27,"remaining_time":1123.525848,"learn":[2.908119111]},{"test":[2.11308851],"passed_time":33.41445608,"iteration":28,"remaining_time":1118.808167,"learn":[2.905779574]},{"test":[2.111107625],"passed_time":34.49309331,"iteration":29,"remaining_time":1115.276684,"learn":[2.903732691]},{"test":[2.111193692],"passed_time":35.74198949,"iteration":30,"remaining_time":1117.225414,"learn":[2.901695458]},{"test":[2.112416843],"passed_time":36.85131891,"iteration":31,"remaining_time":1114.752397,"learn":[2.900773461]},{"test":[2.111082723],"passed_time":38.00641278,"iteration":32,"remaining_time":1113.703066,"learn":[2.899601112]},{"test":[2.109709265],"passed_time":39.07615528,"iteration":33,"remaining_time":1110.222529,"learn":[2.897014321]},{"test":[2.110406818],"passed_time":40.18652278,"iteration":34,"remaining_time":1107.999842,"learn":[2.895424933]},{"test":[2.109771925],"passed_time":41.4511565,"iteration":35,"remaining_time":1109.969857,"learn":[2.894531394]},{"test":[2.107890541],"passed_time":42.49188033,"iteration":36,"remaining_time":1105.937318,"learn":[2.891969821]},{"test":[2.108361277],"passed_time":43.57790396,"iteration":37,"remaining_time":1103.209042,"learn":[2.891462896]},{"test":[2.108179244],"passed_time":44.74067623,"iteration":38,"remaining_time":1102.45615,"learn":[2.890179282]},{"test":[2.107054693],"passed_time":45.85389021,"iteration":39,"remaining_time":1100.493365,"learn":[2.88754129]},{"test":[2.105661336],"passed_time":46.88873392,"iteration":40,"remaining_time":1096.738923,"learn":[2.885325791]},{"test":[2.105056213],"passed_time":47.92698829,"iteration":41,"remaining_time":1093.191781,"learn":[2.884113903]},{"test":[2.104382435],"passed_time":49.16422122,"iteration":42,"remaining_time":1094.189761,"learn":[2.88197285]},{"test":[2.104852887],"passed_time":50.23966104,"iteration":43,"remaining_time":1091.570817,"learn":[2.879917987]},{"test":[2.104403533],"passed_time":51.15981259,"iteration":44,"remaining_time":1085.724912,"learn":[2.878939147]},{"test":[2.105274794],"passed_time":52.33635857,"iteration":45,"remaining_time":1085.410567,"learn":[2.87852936]},{"test":[2.105548934],"passed_time":53.49059953,"iteration":46,"remaining_time":1084.607263,"learn":[2.877941717]},{"test":[2.105043258],"passed_time":54.57983996,"iteration":47,"remaining_time":1082.500159,"learn":[2.87746067]},{"test":[2.104956449],"passed_time":55.83372993,"iteration":48,"remaining_time":1083.630146,"learn":[2.876623448]},{"test":[2.105546876],"passed_time":57.10476452,"iteration":49,"remaining_time":1084.990526,"learn":[2.875301233]},{"test":[2.103148831],"passed_time":58.3427937,"iteration":50,"remaining_time":1085.633553,"learn":[2.873830119]},{"test":[2.104818778],"passed_time":59.52232094,"iteration":51,"remaining_time":1085.137697,"learn":[2.872600524]},{"test":[2.10201387],"passed_time":60.46722138,"iteration":52,"remaining_time":1080.423748,"learn":[2.871625145]},{"test":[2.103448365],"passed_time":61.65462333,"iteration":53,"remaining_time":1080.097661,"learn":[2.870154091]},{"test":[2.10258366],"passed_time":62.79310845,"iteration":54,"remaining_time":1078.899773,"learn":[2.869430296]},{"test":[2.101061141],"passed_time":63.79082747,"iteration":55,"remaining_time":1075.331092,"learn":[2.868632921]},{"test":[2.103613961],"passed_time":64.82138386,"iteration":56,"remaining_time":1072.395877,"learn":[2.867326165]},{"test":[2.102984604],"passed_time":65.9059477,"iteration":57,"remaining_time":1070.403495,"learn":[2.866166616]},{"test":[2.102195391],"passed_time":66.75235077,"iteration":58,"remaining_time":1064.643425,"learn":[2.865500878]},{"test":[2.100512804],"passed_time":67.90753602,"iteration":59,"remaining_time":1063.884731,"learn":[2.863884285]},{"test":[2.098704105],"passed_time":69.02798985,"iteration":60,"remaining_time":1062.578401,"learn":[2.861299011]},{"test":[2.098482038],"passed_time":70.11414705,"iteration":61,"remaining_time":1060.759192,"learn":[2.860416229]},{"test":[2.097960782],"passed_time":71.2394849,"iteration":62,"remaining_time":1059.54599,"learn":[2.859726887]},{"test":[2.096722417],"passed_time":72.49928015,"iteration":63,"remaining_time":1060.301972,"learn":[2.858924593]},{"test":[2.100632001],"passed_time":73.58744616,"iteration":64,"remaining_time":1058.52711,"learn":[2.85776785]},{"test":[2.100380659],"passed_time":74.65474411,"iteration":65,"remaining_time":1056.477742,"learn":[2.857156382]},{"test":[2.100420671],"passed_time":75.60365807,"iteration":66,"remaining_time":1052.809149,"learn":[2.856820262]},{"test":[2.09742634],"passed_time":76.8614458,"iteration":67,"remaining_time":1053.453934,"learn":[2.854780069]},{"test":[2.098610048],"passed_time":77.94690457,"iteration":68,"remaining_time":1051.718379,"learn":[2.853647035]},{"test":[2.098745287],"passed_time":79.21564633,"iteration":69,"remaining_time":1052.436444,"learn":[2.852327442]},{"test":[2.098694066],"passed_time":80.30976317,"iteration":70,"remaining_time":1050.813662,"learn":[2.852152466]},{"test":[2.099006344],"passed_time":81.50168049,"iteration":71,"remaining_time":1050.466104,"learn":[2.851938976]},{"test":[2.09916382],"passed_time":82.68282571,"iteration":72,"remaining_time":1049.958622,"learn":[2.851529067]},{"test":[2.098986618],"passed_time":83.97364219,"iteration":73,"remaining_time":1050.805306,"learn":[2.850693195]}]},"total_iterations":1000,"passed_iterations":73},"path":"catboost_info","name":"catboost_info"}},"_view_module":"catboost-widget","_model_module_version":"^1.0.0","_view_count":null,"_view_module_version":"^1.0.0","layout":"IPY_MODEL_fffe49ef8b6f4db8bbdfd007bdfc68e8","_model_module":"catboost-widget"}}}}},"cells":[{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"QbTvN4p4nw3p","executionInfo":{"status":"ok","timestamp":1632704683885,"user_tz":-540,"elapsed":24361,"user":{"displayName":"中田光隆","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"06690068059315868411"}},"outputId":"5869ec51-f570-4c63-dac2-b6908995517e"},"source":["!pip install signate\n","\n","from googleapiclient.discovery import build\n","import io, os\n","from googleapiclient.http import MediaIoBaseDownload\n","from google.colab import auth\n","\n","\n","auth.authenticate_user()\n","\n","drive_service = build('drive', 'v3')\n","results = drive_service.files().list(\n","        q=\"name = 'signate.json'\", fields=\"files(id)\").execute()\n","signate_api_key = results.get('files', [])\n","\n","filename = \"/root/.signate/signate.json\"\n","os.makedirs(os.path.dirname(filename), exist_ok=True)\n","\n","request = drive_service.files().get_media(fileId=signate_api_key[0]['id'])\n","fh = io.FileIO(filename, 'wb')\n","downloader = MediaIoBaseDownload(fh, request)\n","done = False\n","while done is False:\n","    status, done = downloader.next_chunk()\n","    print(\"Download %d%%.\" % int(status.progress() * 100))\n","os.chmod(filename, 600)"],"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Collecting signate\n","  Downloading signate-0.9.8-py3-none-any.whl (37 kB)\n","Requirement already satisfied: certifi in /usr/local/lib/python3.7/dist-packages (from signate) (2021.5.30)\n","Requirement already satisfied: python-dateutil in /usr/local/lib/python3.7/dist-packages (from signate) (2.8.2)\n","Collecting urllib3>=1.26.4\n","  Downloading urllib3-1.26.7-py2.py3-none-any.whl (138 kB)\n","\u001b[K     |████████████████████████████████| 138 kB 8.6 MB/s \n","\u001b[?25hCollecting wget\n","  Downloading wget-3.2.zip (10 kB)\n","Requirement already satisfied: click in /usr/local/lib/python3.7/dist-packages (from signate) (7.1.2)\n","Requirement already satisfied: tabulate in /usr/local/lib/python3.7/dist-packages (from signate) (0.8.9)\n","Requirement already satisfied: six>=1.10 in /usr/local/lib/python3.7/dist-packages (from signate) (1.15.0)\n","Building wheels for collected packages: wget\n","  Building wheel for wget (setup.py) ... \u001b[?25l\u001b[?25hdone\n","  Created wheel for wget: filename=wget-3.2-py3-none-any.whl size=9672 sha256=16690645683bdc8cd3fece267dc335ffcb92422bfca8bf5785497c8ad0c7f484\n","  Stored in directory: /root/.cache/pip/wheels/a1/b6/7c/0e63e34eb06634181c63adacca38b79ff8f35c37e3c13e3c02\n","Successfully built wget\n","Installing collected packages: wget, urllib3, signate\n","  Attempting uninstall: urllib3\n","    Found existing installation: urllib3 1.24.3\n","    Uninstalling urllib3-1.24.3:\n","      Successfully uninstalled urllib3-1.24.3\n","\u001b[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\n","requests 2.23.0 requires urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1, but you have urllib3 1.26.7 which is incompatible.\n","datascience 0.10.6 requires folium==0.2.1, but you have folium 0.8.3 which is incompatible.\u001b[0m\n","Successfully installed signate-0.9.8 urllib3-1.26.7 wget-3.2\n"]},{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.7/dist-packages/requests/__init__.py:91: RequestsDependencyWarning: urllib3 (1.26.7) or chardet (3.0.4) doesn't match a supported version!\n","  RequestsDependencyWarning)\n"]},{"output_type":"stream","name":"stdout","text":["Download 100%.\n"]}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"0KR418EDn8O5","executionInfo":{"status":"ok","timestamp":1632704700964,"user_tz":-540,"elapsed":17084,"user":{"displayName":"中田光隆","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"06690068059315868411"}},"outputId":"67e535a8-0354-49fa-e47b-d85f97107908"},"source":["! signate list\n","! signate files --competition-id=492\n","! signate download --competition-id=492"],"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["  competitionId  title                                                                  closing     prize            submitters\n","---------------  ---------------------------------------------------------------------  ----------  -------------  ------------\n","              1  【練習問題】銀行の顧客ターゲティング                                   -                                  5139\n","             24  【練習問題】お弁当の需要予測                                           -                                  6328\n","             27  【練習問題】Jリーグの観客動員数予測                                    -                                  1432\n","            100  【練習問題】手書き文字認識                                             -           Knowledge               166\n","            102  【練習問題】タイタニックの生存予測                                     -           Knowledge              1055\n","            103  【練習問題】音楽ラベリング                                             -           Knowledge                60\n","            104  【練習問題】スパムメール分類                                           -           Knowledge               129\n","            105  【練習問題】毒キノコの分類                                             -           Knowledge               244\n","            106  【練習問題】アワビの年齢予測                                           -           Knowledge               352\n","            107  【練習問題】国勢調査からの収入予測                                     -           Knowledge               521\n","            108  【練習問題】画像ラベリング（20種類）                                   -           Knowledge               220\n","            112  【練習問題】ワインの品種の予測                                         -           Knowledge               296\n","            113  【練習問題】山火事の消失面積予測                                       -           Knowledge               185\n","            114  【練習問題】レンタル自転車の利用者数予測                               -           Knowledge               368\n","            115  【練習問題】アヤメの分類                                               -           Knowledge               333\n","            116  【練習問題】活動センサーログからの動作予測                             -           Knowledge                42\n","            118  【練習問題】テニスの試合結果の予測                                     -           Knowledge               108\n","            121  【練習問題】自動車の走行距離予測                                       -           Knowledge              1319\n","            122  【練習問題】自動車の評価                                               -           Knowledge               302\n","            123  【練習問題】オゾンレベルの分類                                         -           Knowledge                35\n","            124  【練習問題】ボットの判別                                               -           Knowledge               287\n","            125  【練習問題】ガラスの分類                                               -           Knowledge               125\n","            126  【練習問題】林型の分類                                                 -           Knowledge                26\n","            127  【練習問題】ゲーム選手のリーグ分類                                     -           Knowledge                35\n","            128  【練習問題】ステンレス板の欠陥分類                                     -           Knowledge                42\n","            129  【練習問題】都市サイクルの燃料消費量予測                               -           Knowledge                68\n","            130  【練習問題】天秤のバランス分類                                         -           Knowledge                53\n","            132  【練習問題】ネット広告のクリック予測                                   -           Knowledge               140\n","            133  【練習問題】画像ラベリング（10種類）                                   -                                   330\n","            135  【練習問題】ネット画像の分類                                           -                                    71\n","            262  【SOTA】国立国会図書館の画像データレイアウト認識                       2100-12-31  -                        55\n","            263  【SOTA】産業技術総合研究所 衛星画像分析コンテスト                      2100-12-31  -                        32\n","            264  【SOTA】マイナビ × SIGNATE Student Cup 2019: 賃貸物件の家賃予測        2100-12-31  -                       317\n","            265  【練習問題】健診データによる肝疾患判定                                 -                                   443\n","            266  【練習問題】民泊サービスの宿泊価格予測                                 -                                   457\n","            267  【SOTA】海洋研究開発機構 熱帯低気圧（台風等）検出アルゴリズム作成      2100-12-31  -                        13\n","            268  【SOTA】オプト レコメンドエンジン作成                                  2100-12-31                           27\n","            269  【SOTA】アップル 引越し需要予測                                        2100-12-31  -                       264\n","            270  【SOTA】Weather Challenge：雲画像予測                                  2100-12-31  -                         5\n","            271  【SOTA】JR西日本 走行中の北陸新幹線車両台車部の着雪量予測              2100-12-31  -                        12\n","            288  【SOTA】Sansan 名刺の項目予測                                          2100-12-31  -                        21\n","            294  【練習問題】債務不履行リスクの低減                                     -                                   135\n","            358  【練習問題】機械稼働音の異常検知                                       -                                    52\n","            404  【練習問題】モノクロ顔画像の感情分類                                   -                                    40\n","            406  【練習問題】鋳造製品の欠陥検出                                         -                                   123\n","            409  【練習問題】株価の推移予測                                             -                                   133\n","            471  医学論文の自動仕分けチャレンジ                                         2021-10-04  総額￥400,000           590\n","            489  【AIQuest2021】PBL_01 需要予測・在庫最適化（小売業）ビジネス導入課題①  2022-02-05                          519\n","            490  【AIQuest2021】PBL_01 需要予測・在庫最適化（小売業）ビジネス導入課題②  2022-02-05                          514\n","            491  【AIQuest2021】PBL_01 需要予測・在庫最適化（小売業）ビジネス導入課題③  2022-02-05                          512\n","            492  【AIQuest2021】PBL_01 需要予測・在庫最適化（小売業）AI課題             2022-02-05                          273\n","  fileId  name                                           title                                                   size  updated_at\n","--------  ---------------------------------------------  --------------------------------------------------  --------  -------------------\n","    2201  PBL01_h_exercise4_question.pdf                 PBL01_需要予測・在庫最適化_h.演習④設問                734746  2021-08-20 09:33:52\n","    2202  PBL01_i_exercise4_supplementary-documents.pdf  PBL01_需要予測・在庫最適化_i.演習④ガイドコンテンツ   1410304  2021-08-19 14:50:11\n","    2114  sales_history.csv                              販売実績データ                                      33293020  2021-08-19 14:50:11\n","    2115  item_categories.csv                            商品カテゴリーデータ                                  113142  2021-08-19 14:50:07\n","    2116  category_names.csv                             カテゴリ名称データ                                       879  2021-08-19 14:50:07\n","    2117  test.csv                                       評価用データ                                           31978  2021-08-19 14:50:07\n","    2118  sample_submission.csv                          応募用サンプルファイル                                 26430  2021-08-19 14:50:07\n","    2410  PBL01_sample_code.ipynb                        サンプルコード（9/18追加）                            107000  2021-09-18 11:24:06\n","    2411  requirements.txt                               サンプルコードについて（9/18追加）                        82  2021-09-18 11:24:30\n","requirements.txt\n","\n","category_names.csv\n","\n","sample_submission.csv\n","\n","test.csv\n","\n","PBL01_sample_code.ipynb\n","\n","item_categories.csv\n","\n","PBL01_h_exercise4_question.pdf\n","\n","PBL01_i_exercise4_supplementary-documents.pdf\n","\n","sales_history.csv\n","\n","\u001b[32m\n","Download completed.\u001b[0m\n"]}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"8nCDxjlIoQJt","executionInfo":{"status":"ok","timestamp":1632704700965,"user_tz":-540,"elapsed":20,"user":{"displayName":"中田光隆","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"06690068059315868411"}},"outputId":"da6334e6-88da-4ae3-b9f6-2e681c7af237"},"source":["# ディレクトリ内のデータの確認\n","%ls"],"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["adc.json                                       requirements.txt\n","category_names.csv                             sales_history.csv\n","item_categories.csv                            \u001b[0m\u001b[01;34msample_data\u001b[0m/\n","PBL01_h_exercise4_question.pdf                 sample_submission.csv\n","PBL01_i_exercise4_supplementary-documents.pdf  test.csv\n","PBL01_sample_code.ipynb\n"]}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Xtt3uuaQmXId","executionInfo":{"status":"ok","timestamp":1632704717727,"user_tz":-540,"elapsed":16776,"user":{"displayName":"中田光隆","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"06690068059315868411"}},"outputId":"8eae7e2c-6ef1-489c-8198-34a4175db137"},"source":["# Googleドライブのマウント\n","from google.colab import drive\n","drive.mount('/content/drive/')"],"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Mounted at /content/drive/\n"]}]},{"cell_type":"markdown","metadata":{"id":"RnctI5aCzH37"},"source":["#baseline"]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"FXOPTl2QySj6","executionInfo":{"status":"ok","timestamp":1632704821468,"user_tz":-540,"elapsed":6517,"user":{"displayName":"中田光隆","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"06690068059315868411"}},"outputId":"ef8d8bae-6731-430a-96de-585abbfa4d79"},"source":["!pip install catboost"],"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Requirement already satisfied: catboost in /usr/local/lib/python3.7/dist-packages (0.26.1)\n","Requirement already satisfied: matplotlib in /usr/local/lib/python3.7/dist-packages (from catboost) (3.2.2)\n","Requirement already satisfied: scipy in /usr/local/lib/python3.7/dist-packages (from catboost) (1.4.1)\n","Requirement already satisfied: pandas>=0.24.0 in /usr/local/lib/python3.7/dist-packages (from catboost) (1.1.5)\n","Requirement already satisfied: numpy>=1.16.0 in /usr/local/lib/python3.7/dist-packages (from catboost) (1.19.5)\n","Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from catboost) (1.15.0)\n","Requirement already satisfied: plotly in /usr/local/lib/python3.7/dist-packages (from catboost) (4.4.1)\n","Requirement already satisfied: graphviz in /usr/local/lib/python3.7/dist-packages (from catboost) (0.10.1)\n","Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.24.0->catboost) (2018.9)\n","Requirement already satisfied: python-dateutil>=2.7.3 in /usr/local/lib/python3.7/dist-packages (from pandas>=0.24.0->catboost) (2.8.2)\n","Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->catboost) (1.3.2)\n","Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.7/dist-packages (from matplotlib->catboost) (0.10.0)\n","Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.7/dist-packages (from matplotlib->catboost) (2.4.7)\n","Requirement already satisfied: retrying>=1.3.3 in /usr/local/lib/python3.7/dist-packages (from plotly->catboost) (1.3.3)\n","Requirement already satisfied: ipywidgets in /usr/local/lib/python3.7/dist-packages (7.6.5)\n","Requirement already satisfied: nbformat>=4.2.0 in /usr/local/lib/python3.7/dist-packages (from ipywidgets) (5.1.3)\n","Requirement already satisfied: ipython>=4.0.0 in /usr/local/lib/python3.7/dist-packages (from ipywidgets) (5.5.0)\n","Requirement already satisfied: ipykernel>=4.5.1 in /usr/local/lib/python3.7/dist-packages (from ipywidgets) (4.10.1)\n","Requirement already satisfied: widgetsnbextension~=3.5.0 in /usr/local/lib/python3.7/dist-packages (from ipywidgets) (3.5.1)\n","Requirement already satisfied: traitlets>=4.3.1 in /usr/local/lib/python3.7/dist-packages (from ipywidgets) (5.1.0)\n","Requirement already satisfied: jupyterlab-widgets>=1.0.0 in /usr/local/lib/python3.7/dist-packages (from ipywidgets) (1.0.2)\n","Requirement already satisfied: ipython-genutils~=0.2.0 in /usr/local/lib/python3.7/dist-packages (from ipywidgets) (0.2.0)\n","Requirement already satisfied: jupyter-client in /usr/local/lib/python3.7/dist-packages (from ipykernel>=4.5.1->ipywidgets) (5.3.5)\n","Requirement already satisfied: tornado>=4.0 in /usr/local/lib/python3.7/dist-packages (from ipykernel>=4.5.1->ipywidgets) (5.1.1)\n","Requirement already satisfied: simplegeneric>0.8 in /usr/local/lib/python3.7/dist-packages (from ipython>=4.0.0->ipywidgets) (0.8.1)\n","Requirement already satisfied: prompt-toolkit<2.0.0,>=1.0.4 in /usr/local/lib/python3.7/dist-packages (from ipython>=4.0.0->ipywidgets) (1.0.18)\n","Requirement already satisfied: pickleshare in /usr/local/lib/python3.7/dist-packages (from ipython>=4.0.0->ipywidgets) (0.7.5)\n","Requirement already satisfied: pexpect in /usr/local/lib/python3.7/dist-packages (from ipython>=4.0.0->ipywidgets) (4.8.0)\n","Requirement already satisfied: setuptools>=18.5 in /usr/local/lib/python3.7/dist-packages (from ipython>=4.0.0->ipywidgets) (57.4.0)\n","Requirement already satisfied: decorator in /usr/local/lib/python3.7/dist-packages (from ipython>=4.0.0->ipywidgets) (4.4.2)\n","Requirement already satisfied: pygments in /usr/local/lib/python3.7/dist-packages (from ipython>=4.0.0->ipywidgets) (2.6.1)\n","Requirement already satisfied: jupyter-core in /usr/local/lib/python3.7/dist-packages (from nbformat>=4.2.0->ipywidgets) (4.7.1)\n","Requirement already satisfied: jsonschema!=2.5.0,>=2.4 in /usr/local/lib/python3.7/dist-packages (from nbformat>=4.2.0->ipywidgets) (2.6.0)\n","Requirement already satisfied: six>=1.9.0 in /usr/local/lib/python3.7/dist-packages (from prompt-toolkit<2.0.0,>=1.0.4->ipython>=4.0.0->ipywidgets) (1.15.0)\n","Requirement already satisfied: wcwidth in /usr/local/lib/python3.7/dist-packages (from prompt-toolkit<2.0.0,>=1.0.4->ipython>=4.0.0->ipywidgets) (0.2.5)\n","Requirement already satisfied: notebook>=4.4.1 in /usr/local/lib/python3.7/dist-packages (from widgetsnbextension~=3.5.0->ipywidgets) (5.3.1)\n","Requirement already satisfied: terminado>=0.8.1 in /usr/local/lib/python3.7/dist-packages (from notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (0.12.1)\n","Requirement already satisfied: jinja2 in /usr/local/lib/python3.7/dist-packages (from notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (2.11.3)\n","Requirement already satisfied: nbconvert in /usr/local/lib/python3.7/dist-packages (from notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (5.6.1)\n","Requirement already satisfied: Send2Trash in /usr/local/lib/python3.7/dist-packages (from notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (1.8.0)\n","Requirement already satisfied: pyzmq>=13 in /usr/local/lib/python3.7/dist-packages (from jupyter-client->ipykernel>=4.5.1->ipywidgets) (22.3.0)\n","Requirement already satisfied: python-dateutil>=2.1 in /usr/local/lib/python3.7/dist-packages (from jupyter-client->ipykernel>=4.5.1->ipywidgets) (2.8.2)\n","Requirement already satisfied: ptyprocess in /usr/local/lib/python3.7/dist-packages (from terminado>=0.8.1->notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (0.7.0)\n","Requirement already satisfied: MarkupSafe>=0.23 in /usr/local/lib/python3.7/dist-packages (from jinja2->notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (2.0.1)\n","Requirement already satisfied: testpath in /usr/local/lib/python3.7/dist-packages (from nbconvert->notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (0.5.0)\n","Requirement already satisfied: mistune<2,>=0.8.1 in /usr/local/lib/python3.7/dist-packages (from nbconvert->notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (0.8.4)\n","Requirement already satisfied: entrypoints>=0.2.2 in /usr/local/lib/python3.7/dist-packages (from nbconvert->notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (0.3)\n","Requirement already satisfied: defusedxml in /usr/local/lib/python3.7/dist-packages (from nbconvert->notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (0.7.1)\n","Requirement already satisfied: pandocfilters>=1.4.1 in /usr/local/lib/python3.7/dist-packages (from nbconvert->notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (1.5.0)\n","Requirement already satisfied: bleach in /usr/local/lib/python3.7/dist-packages (from nbconvert->notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (4.1.0)\n","Requirement already satisfied: packaging in /usr/local/lib/python3.7/dist-packages (from bleach->nbconvert->notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (21.0)\n","Requirement already satisfied: webencodings in /usr/local/lib/python3.7/dist-packages (from bleach->nbconvert->notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (0.5.1)\n","Requirement already satisfied: pyparsing>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging->bleach->nbconvert->notebook>=4.4.1->widgetsnbextension~=3.5.0->ipywidgets) (2.4.7)\n"]}]},{"cell_type":"code","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000,"referenced_widgets":["f50370b6d5624e099320a5cba46ebc1f"]},"id":"t1t8jDKrp4ah","executionInfo":{"status":"ok","timestamp":1632705123186,"user_tz":-540,"elapsed":91225,"user":{"displayName":"中田光隆","photoUrl":"https://lh3.googleusercontent.com/a/default-user=s64","userId":"06690068059315868411"}},"outputId":"a2f09b04-6e62-458f-aa29-0fa23e0ea232"},"source":["'''\n","初期設定\n","'''\n","\n","DATA_PATH = '/content/drive/My Drive/AIQuest2021/PBL_01_需要予測/Phase2/demand/processed_data/processed_train_test_df.csv'\n","SAMPLESUB_PATH = 'sample_submission.csv'\n","SUB_PATH = '/content/drive/My Drive/AIQuest2021/PBL_01_需要予測/Phase2/demand/submit/demand_CatBoost.csv'\n","\n","\n","# ライブラリのインポート\n","import pandas as pd\n","import numpy as np\n","import matplotlib.pyplot as plt\n","import seaborn as sns\n","from sklearn.preprocessing import LabelEncoder\n","\n","# データの読み込み\n","df = pd.read_csv(DATA_PATH)\n","print(df.isnull().sum())\n","print(df.dtypes)\n","\n","# object型の変数の取得\n","categories = df.columns[df.dtypes == 'object']\n","print(categories)\n","\n","# label Encoding\n","for cat in categories:\n","    le = LabelEncoder() \n","    print(cat)\n","    \n","    df[cat].fillna('missing', inplace=True)\n","    le = le.fit(df[cat])\n","    df[cat] = le.transform(df[cat])\n","    # LabelEncoderは数値に変換するだけであるため、最後にastype('category')としておく\n","    df[cat] = df[cat].astype('category') \n","\n","# trainとtestに分割\n","train = df[~df['sales'].isnull()]\n","test = df[df['sales'].isnull()]\n","\n","# 説明変数と目的変数を指定\n","X_train = train.drop(['sales'], axis=1)\n","Y_train = train['sales']\n","X_test = test.drop(['sales'], axis=1)\n","\n","'''\n","モデルの構築と評価\n","'''\n","\n","# ライブラリのインポート\n","from catboost import CatBoostRegressor\n","from catboost import Pool\n","from sklearn.model_selection import train_test_split\n","from sklearn.metrics import mean_squared_error\n","\n","# カテゴリカル変数の指定\n","categorical_features = np.where(X_train.dtypes != np.float)[0]\n","\n","x_train, x_valid, y_train, y_valid = train_test_split(X_train, Y_train,\n","                                                      test_size=0.3, random_state=0,\n","                                                      shuffle=False)\n","\n","train_pool = Pool(x_train, y_train, cat_features=categorical_features)\n","valid_pool = Pool(x_valid, y_valid, cat_features=categorical_features)   \n","\n","# パラメーターの設定\n","params = {\n","    'loss_function': 'RMSE',\n","    'num_boost_round': 1000,\n","    'early_stopping_rounds': 10,\n","    'random_state': 42,    \n","}\n","\n","model = CatBoostRegressor(**params)\n","model.fit(train_pool, \n","          eval_set=valid_pool,    # 検証用データ\n","          use_best_model=True,       # 最も精度が高かったモデルを使用するかの設定\n","          plot=True)\n","    \n","y_pred = model.predict(valid_pool)\n","rmse = np.sqrt(mean_squared_error(y_valid, y_pred))\n","print(rmse)\n","\n","# 特徴量重要度の表示\n","df_feature_importance = pd.DataFrame(model.get_feature_importance(prettified=True))\n","plt.figure(figsize=(12, 6))\n","feature_plot= sns.barplot(x=\"Importances\", y=\"Feature Id\", data=df_feature_importance,palette=\"cool\")\n","plt.title('features importance');\n","\n","\"\"\"\n","予測精度：\n","2.096722417169216\n","\"\"\"\n","\n","'''\n","テストデータの予測\n","'''\n","\n","# テストデータにおける予測\n","pred = model.predict(X_test)\n","\n","'''\n","提出\n","'''\n","\n","# 提出用サンプルの読み込み\n","sub = pd.read_csv(SAMPLESUB_PATH, header=None)\n","\n","# カラム1の値を置き換え\n","sub[1] = pred\n","\n","# CSVファイルの出力\n","sub.to_csv(SUB_PATH, header=None, index=False)\n","\n","\"\"\"\n","スコア：\n","3.8447951\n","\"\"\""],"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["month_bining          0\n","id                    0\n","store_id              0\n","sales              3060\n","sales_before_12       0\n","sales_before_11       0\n","sales_before_10       0\n","sales_before_9        0\n","sales_before_8        0\n","sales_before_7        0\n","sales_before_6        0\n","sales_before_5        0\n","sales_before_4        0\n","sales_before_3        0\n","sales_before_2        0\n","category_id           0\n","category              0\n","category_detail       0\n","year                  0\n","month                 0\n","dtype: int64\n","month_bining       float64\n","id                 float64\n","store_id           float64\n","sales              float64\n","sales_before_12    float64\n","sales_before_11    float64\n","sales_before_10    float64\n","sales_before_9     float64\n","sales_before_8     float64\n","sales_before_7     float64\n","sales_before_6     float64\n","sales_before_5     float64\n","sales_before_4     float64\n","sales_before_3     float64\n","sales_before_2     float64\n","category_id          int64\n","category            object\n","category_detail     object\n","year                 int64\n","month                int64\n","dtype: object\n","Index(['category', 'category_detail'], dtype='object')\n","category\n","category_detail\n"]},{"output_type":"stream","name":"stderr","text":["Custom logger is already specified. Specify more than one logger at same time is not thread safe."]},{"output_type":"display_data","data":{"application/vnd.jupyter.widget-view+json":{"model_id":"f50370b6d5624e099320a5cba46ebc1f","version_minor":0,"version_major":2},"text/plain":["MetricVisualizer(layout=Layout(align_self='stretch', height='500px'))"]},"metadata":{}},{"output_type":"stream","name":"stdout","text":["Learning rate set to 0.174687\n","0:\tlearn: 3.1260342\ttest: 2.3073629\tbest: 2.3073629 (0)\ttotal: 1.19s\tremaining: 19m 47s\n","1:\tlearn: 3.0830701\ttest: 2.2504895\tbest: 2.2504895 (1)\ttotal: 2.34s\tremaining: 19m 27s\n","2:\tlearn: 3.0522218\ttest: 2.2151492\tbest: 2.2151492 (2)\ttotal: 3.51s\tremaining: 19m 25s\n","3:\tlearn: 3.0296826\ttest: 2.1866069\tbest: 2.1866069 (3)\ttotal: 4.47s\tremaining: 18m 33s\n","4:\tlearn: 3.0136469\ttest: 2.1707439\tbest: 2.1707439 (4)\ttotal: 5.53s\tremaining: 18m 20s\n","5:\tlearn: 3.0013072\ttest: 2.1590216\tbest: 2.1590216 (5)\ttotal: 6.59s\tremaining: 18m 11s\n","6:\tlearn: 2.9921135\ttest: 2.1521766\tbest: 2.1521766 (6)\ttotal: 7.66s\tremaining: 18m 6s\n","7:\tlearn: 2.9845199\ttest: 2.1458230\tbest: 2.1458230 (7)\ttotal: 8.71s\tremaining: 17m 59s\n","8:\tlearn: 2.9781469\ttest: 2.1426472\tbest: 2.1426472 (8)\ttotal: 9.96s\tremaining: 18m 16s\n","9:\tlearn: 2.9716519\ttest: 2.1403389\tbest: 2.1403389 (9)\ttotal: 11.1s\tremaining: 18m 15s\n","10:\tlearn: 2.9679189\ttest: 2.1373448\tbest: 2.1373448 (10)\ttotal: 12.3s\tremaining: 18m 28s\n","11:\tlearn: 2.9593183\ttest: 2.1265465\tbest: 2.1265465 (11)\ttotal: 13.6s\tremaining: 18m 41s\n","12:\tlearn: 2.9532268\ttest: 2.1217328\tbest: 2.1217328 (12)\ttotal: 14.9s\tremaining: 18m 49s\n","13:\tlearn: 2.9490105\ttest: 2.1195012\tbest: 2.1195012 (13)\ttotal: 15.9s\tremaining: 18m 40s\n","14:\tlearn: 2.9452948\ttest: 2.1190129\tbest: 2.1190129 (14)\ttotal: 16.9s\tremaining: 18m 29s\n","15:\tlearn: 2.9414205\ttest: 2.1189779\tbest: 2.1189779 (15)\ttotal: 18s\tremaining: 18m 26s\n","16:\tlearn: 2.9393109\ttest: 2.1194415\tbest: 2.1189779 (15)\ttotal: 19.1s\tremaining: 18m 24s\n","17:\tlearn: 2.9369326\ttest: 2.1184446\tbest: 2.1184446 (17)\ttotal: 20.3s\tremaining: 18m 25s\n","18:\tlearn: 2.9318341\ttest: 2.1152524\tbest: 2.1152524 (18)\ttotal: 21.6s\tremaining: 18m 32s\n","19:\tlearn: 2.9279742\ttest: 2.1152484\tbest: 2.1152484 (19)\ttotal: 22.8s\tremaining: 18m 38s\n","20:\tlearn: 2.9253867\ttest: 2.1169303\tbest: 2.1152484 (19)\ttotal: 24s\tremaining: 18m 39s\n","21:\tlearn: 2.9233695\ttest: 2.1154918\tbest: 2.1152484 (19)\ttotal: 25.3s\tremaining: 18m 46s\n","22:\tlearn: 2.9189787\ttest: 2.1170381\tbest: 2.1152484 (19)\ttotal: 26.6s\tremaining: 18m 50s\n","23:\tlearn: 2.9149547\ttest: 2.1167377\tbest: 2.1152484 (19)\ttotal: 27.8s\tremaining: 18m 52s\n","24:\tlearn: 2.9136566\ttest: 2.1157807\tbest: 2.1152484 (19)\ttotal: 29s\tremaining: 18m 51s\n","25:\tlearn: 2.9114926\ttest: 2.1156346\tbest: 2.1152484 (19)\ttotal: 30s\tremaining: 18m 45s\n","26:\tlearn: 2.9098061\ttest: 2.1133143\tbest: 2.1133143 (26)\ttotal: 31.3s\tremaining: 18m 46s\n","27:\tlearn: 2.9081191\ttest: 2.1138709\tbest: 2.1133143 (26)\ttotal: 32.4s\tremaining: 18m 43s\n","28:\tlearn: 2.9057796\ttest: 2.1130885\tbest: 2.1130885 (28)\ttotal: 33.4s\tremaining: 18m 38s\n","29:\tlearn: 2.9037327\ttest: 2.1111076\tbest: 2.1111076 (29)\ttotal: 34.5s\tremaining: 18m 35s\n","30:\tlearn: 2.9016955\ttest: 2.1111937\tbest: 2.1111076 (29)\ttotal: 35.7s\tremaining: 18m 37s\n","31:\tlearn: 2.9007735\ttest: 2.1124168\tbest: 2.1111076 (29)\ttotal: 36.9s\tremaining: 18m 34s\n","32:\tlearn: 2.8996011\ttest: 2.1110827\tbest: 2.1110827 (32)\ttotal: 38s\tremaining: 18m 33s\n","33:\tlearn: 2.8970143\ttest: 2.1097093\tbest: 2.1097093 (33)\ttotal: 39.1s\tremaining: 18m 30s\n","34:\tlearn: 2.8954249\ttest: 2.1104068\tbest: 2.1097093 (33)\ttotal: 40.2s\tremaining: 18m 28s\n","35:\tlearn: 2.8945314\ttest: 2.1097719\tbest: 2.1097093 (33)\ttotal: 41.5s\tremaining: 18m 29s\n","36:\tlearn: 2.8919698\ttest: 2.1078905\tbest: 2.1078905 (36)\ttotal: 42.5s\tremaining: 18m 25s\n","37:\tlearn: 2.8914629\ttest: 2.1083613\tbest: 2.1078905 (36)\ttotal: 43.6s\tremaining: 18m 23s\n","38:\tlearn: 2.8901793\ttest: 2.1081792\tbest: 2.1078905 (36)\ttotal: 44.7s\tremaining: 18m 22s\n","39:\tlearn: 2.8875413\ttest: 2.1070547\tbest: 2.1070547 (39)\ttotal: 45.9s\tremaining: 18m 20s\n","40:\tlearn: 2.8853258\ttest: 2.1056613\tbest: 2.1056613 (40)\ttotal: 46.9s\tremaining: 18m 16s\n","41:\tlearn: 2.8841139\ttest: 2.1050562\tbest: 2.1050562 (41)\ttotal: 47.9s\tremaining: 18m 13s\n","42:\tlearn: 2.8819728\ttest: 2.1043824\tbest: 2.1043824 (42)\ttotal: 49.2s\tremaining: 18m 14s\n","43:\tlearn: 2.8799180\ttest: 2.1048529\tbest: 2.1043824 (42)\ttotal: 50.2s\tremaining: 18m 11s\n","44:\tlearn: 2.8789391\ttest: 2.1044035\tbest: 2.1043824 (42)\ttotal: 51.2s\tremaining: 18m 5s\n","45:\tlearn: 2.8785294\ttest: 2.1052748\tbest: 2.1043824 (42)\ttotal: 52.3s\tremaining: 18m 5s\n","46:\tlearn: 2.8779417\ttest: 2.1055489\tbest: 2.1043824 (42)\ttotal: 53.5s\tremaining: 18m 4s\n","47:\tlearn: 2.8774607\ttest: 2.1050433\tbest: 2.1043824 (42)\ttotal: 54.6s\tremaining: 18m 2s\n","48:\tlearn: 2.8766234\ttest: 2.1049564\tbest: 2.1043824 (42)\ttotal: 55.8s\tremaining: 18m 3s\n","49:\tlearn: 2.8753012\ttest: 2.1055469\tbest: 2.1043824 (42)\ttotal: 57.1s\tremaining: 18m 4s\n","50:\tlearn: 2.8738301\ttest: 2.1031488\tbest: 2.1031488 (50)\ttotal: 58.3s\tremaining: 18m 5s\n","51:\tlearn: 2.8726005\ttest: 2.1048188\tbest: 2.1031488 (50)\ttotal: 59.5s\tremaining: 18m 5s\n","52:\tlearn: 2.8716251\ttest: 2.1020139\tbest: 2.1020139 (52)\ttotal: 1m\tremaining: 18m\n","53:\tlearn: 2.8701541\ttest: 2.1034484\tbest: 2.1020139 (52)\ttotal: 1m 1s\tremaining: 18m\n","54:\tlearn: 2.8694303\ttest: 2.1025837\tbest: 2.1020139 (52)\ttotal: 1m 2s\tremaining: 17m 58s\n","55:\tlearn: 2.8686329\ttest: 2.1010611\tbest: 2.1010611 (55)\ttotal: 1m 3s\tremaining: 17m 55s\n","56:\tlearn: 2.8673262\ttest: 2.1036140\tbest: 2.1010611 (55)\ttotal: 1m 4s\tremaining: 17m 52s\n","57:\tlearn: 2.8661666\ttest: 2.1029846\tbest: 2.1010611 (55)\ttotal: 1m 5s\tremaining: 17m 50s\n","58:\tlearn: 2.8655009\ttest: 2.1021954\tbest: 2.1010611 (55)\ttotal: 1m 6s\tremaining: 17m 44s\n","59:\tlearn: 2.8638843\ttest: 2.1005128\tbest: 2.1005128 (59)\ttotal: 1m 7s\tremaining: 17m 43s\n","60:\tlearn: 2.8612990\ttest: 2.0987041\tbest: 2.0987041 (60)\ttotal: 1m 9s\tremaining: 17m 42s\n","61:\tlearn: 2.8604162\ttest: 2.0984820\tbest: 2.0984820 (61)\ttotal: 1m 10s\tremaining: 17m 40s\n","62:\tlearn: 2.8597269\ttest: 2.0979608\tbest: 2.0979608 (62)\ttotal: 1m 11s\tremaining: 17m 39s\n","63:\tlearn: 2.8589246\ttest: 2.0967224\tbest: 2.0967224 (63)\ttotal: 1m 12s\tremaining: 17m 40s\n","64:\tlearn: 2.8577679\ttest: 2.1006320\tbest: 2.0967224 (63)\ttotal: 1m 13s\tremaining: 17m 38s\n","65:\tlearn: 2.8571564\ttest: 2.1003807\tbest: 2.0967224 (63)\ttotal: 1m 14s\tremaining: 17m 36s\n","66:\tlearn: 2.8568203\ttest: 2.1004207\tbest: 2.0967224 (63)\ttotal: 1m 15s\tremaining: 17m 32s\n","67:\tlearn: 2.8547801\ttest: 2.0974263\tbest: 2.0967224 (63)\ttotal: 1m 16s\tremaining: 17m 33s\n","68:\tlearn: 2.8536470\ttest: 2.0986100\tbest: 2.0967224 (63)\ttotal: 1m 17s\tremaining: 17m 31s\n","69:\tlearn: 2.8523274\ttest: 2.0987453\tbest: 2.0967224 (63)\ttotal: 1m 19s\tremaining: 17m 32s\n","70:\tlearn: 2.8521525\ttest: 2.0986941\tbest: 2.0967224 (63)\ttotal: 1m 20s\tremaining: 17m 30s\n","71:\tlearn: 2.8519390\ttest: 2.0990063\tbest: 2.0967224 (63)\ttotal: 1m 21s\tremaining: 17m 30s\n","72:\tlearn: 2.8515291\ttest: 2.0991638\tbest: 2.0967224 (63)\ttotal: 1m 22s\tremaining: 17m 29s\n","73:\tlearn: 2.8506932\ttest: 2.0989866\tbest: 2.0967224 (63)\ttotal: 1m 23s\tremaining: 17m 30s\n","Stopped by overfitting detector  (10 iterations wait)\n","\n","bestTest = 2.096722417\n","bestIteration = 63\n","\n","Shrink model to first 64 iterations.\n","2.096722417169216\n"]},{"output_type":"execute_result","data":{"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"},"text/plain":["'\\nスコア：\\n3.7497572\\n'"]},"metadata":{},"execution_count":10},{"output_type":"display_data","data":{"image/png":"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\n","text/plain":["<Figure size 864x432 with 1 Axes>"]},"metadata":{"needs_background":"light"}}]}]}