{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# **概要**\n\n※ このNotebookは日本語で記載しています。","metadata":{}},{"cell_type":"markdown","source":"# コンペの内容\n\n- ゲームのシステムログの解析がテーマ\n- ゲーム進行時のクリック挙動とゲーム内のクイズの正答率の関係を予測する\n- データ容量が大きい事も特徴的で限られたリソースの中の効率的な解析が求められる\n\n# 対象となるゲームについて\n\n- 教育ゲームのシステムログ（時系列データ）を扱う\n- 教育ゲーム\n  - ゲームは[ここ](https://jowilder-master.netlify.app/?script_type=original)\n  - ソースは[ここ](https://github.com/fielddaylab/jo_wilder)\n  - クリックのみで進行するゲーム\n  - オブジェクトをクリックしていきヒントを集めながらストーリーを進めていく\n  - たまにクイズが発生し、集めたヒントの中から選択し回答する\n  \n# 提出のルール\n\n- KaggleNotebookを提出する\n- submission.csvは配布されたAPIで生成する\n  - Notebookの中で各手法でモデルを作成したあとに\n  - APIによるiteratorでテストが生成され\n  - APIによるpredictで推論を実行\n  - submission.csvが自動で生成される\n\n# 配布データセットについて\n\n- 時系列データ\n- KaggleのRAM8GBに対してtrain.csvは2GBなので注意\n- train.csv:\n  - 時系列データ\n  - session_id, \n\n\n```tree\n/kaggle/input/predict-student-performance-from-game-play/\n  2218.09 MB : train.csv : 時系列データ\n     4.75 MB : train_labels.csv : \n     0.69 MB : test.csv\n     0.00 MB : sample_submission.csv\n/kaggle/input/predict-student-performance-from-game-play/jo_wilder/\n     0.00 MB : __init__.py\n     0.50 MB : competition.cpython-37m-x86_64-linux-gnu.so\n```","metadata":{}},{"cell_type":"markdown","source":"---\n---\n\n# **各要素の確認**\n\n---\n\n# 配布データの確認","metadata":{}},{"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nfrom pathlib import Path\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    print(f'{dirname}/')\n    for filename in filenames:\n        p = Path(dirname) / filename\n        sz = p.stat().st_size\n        print(f'  {sz/1024/1024:0.2f} MB : {p.name}')","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2023-03-04T01:02:37.124444Z","iopub.execute_input":"2023-03-04T01:02:37.124802Z","iopub.status.idle":"2023-03-04T01:02:37.181407Z","shell.execute_reply.started":"2023-03-04T01:02:37.124775Z","shell.execute_reply":"2023-03-04T01:02:37.180597Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"---\n\n# データセットの確認","metadata":{}},{"cell_type":"markdown","source":"## train.csv\n\n- train.csv\n  - システムログ\n  - session_idがプレイヤーのID\n  - 各プレイヤーの行動ログをindexやelasped_time(経過時間)で分類していく\n  - ゲームの中ではイベントの進行具合を示すlevel(0-22)が用意されている\n  - levelは三段階に分かれており、そのチェックポイントで出題される\n  - questionとlevelの関係はに相当するのかはsample_submission.csvで確認できる","metadata":{}},{"cell_type":"code","source":"%%time\ntrain_df = pd.read_csv('/kaggle/input/predict-student-performance-from-game-play/train.csv')\nprint(f'train.csv : shape={train_df.shape}')","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:02:37.183129Z","iopub.execute_input":"2023-03-04T01:02:37.183700Z","iopub.status.idle":"2023-03-04T01:03:44.136124Z","shell.execute_reply.started":"2023-03-04T01:02:37.183665Z","shell.execute_reply":"2023-03-04T01:03:44.135429Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- 各変数の説明：[公式ルール](https://www.kaggle.com/competitions/predict-student-performance-from-game-play/data)\n- 一部データは空 or n_unique==1","metadata":{}},{"cell_type":"code","source":"cols = train_df.columns\nfor n in range(len(cols)//5):\n    print('\\t'.join(cols[n*5:(n+1)*5]))","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:03:44.137260Z","iopub.execute_input":"2023-03-04T01:03:44.137700Z","iopub.status.idle":"2023-03-04T01:03:44.142505Z","shell.execute_reply.started":"2023-03-04T01:03:44.137674Z","shell.execute_reply":"2023-03-04T01:03:44.141739Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- あるプレイヤーの経過時間とレベルの進行の関係は下の通り\n- checkpointで大きく時間がかかっている様子が確認できる","metadata":{}},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n\nsample_player = train_df.session_id.unique()[0]\ndf_sample = train_df[train_df.session_id==sample_player]\nindex_ckpt = np.where(df_sample.event_name=='checkpoint')\n\n# time vs index, time vs level\nfig, axs = plt.subplots(2,1,sharex=True)\ntime_min = df_sample.elapsed_time/60/1000\naxs[0].plot(time_min, df_sample.index)\naxs[1].plot(time_min, df_sample.level)\nfor i_ckpt in index_ckpt:\n    axs[0].plot([time_min.min(), time_min.max()], [i_ckpt,i_ckpt], ls='dotted')\naxs[1].set_xlabel('elapsed_time [min]')\naxs[0].set_ylabel('index'); axs[1].set_ylabel('level')\nfig.tight_layout()","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:03:44.144518Z","iopub.execute_input":"2023-03-04T01:03:44.144990Z","iopub.status.idle":"2023-03-04T01:03:44.500603Z","shell.execute_reply.started":"2023-03-04T01:03:44.144964Z","shell.execute_reply":"2023-03-04T01:03:44.499762Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## train_labels.csv\n\n- プレイヤーが質問に正解したかどうかのデータセット\n- <#session_id>_q<#question_id> : 1 or 0\n- 一列目をsplitすることで、プレイヤーと質問番号に分類する","metadata":{}},{"cell_type":"code","source":"%%time\ntrain_labels_df = pd.read_csv('/kaggle/input/predict-student-performance-from-game-play/train_labels.csv')\nprint(f'train_labels.csv : shape={train_labels_df.shape}')","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:03:44.501865Z","iopub.execute_input":"2023-03-04T01:03:44.502337Z","iopub.status.idle":"2023-03-04T01:03:44.721721Z","shell.execute_reply.started":"2023-03-04T01:03:44.502304Z","shell.execute_reply":"2023-03-04T01:03:44.720591Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"train_labels_df.head()","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:03:44.722943Z","iopub.execute_input":"2023-03-04T01:03:44.723403Z","iopub.status.idle":"2023-03-04T01:03:44.742472Z","shell.execute_reply.started":"2023-03-04T01:03:44.723376Z","shell.execute_reply":"2023-03-04T01:03:44.741128Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## test.csv\n\n- 実際にはこのテストデータで回答するわけではない\n- 3プレイヤー分のデータが含まれる","metadata":{}},{"cell_type":"code","source":"%%time\ntest_df = pd.read_csv('/kaggle/input/predict-student-performance-from-game-play/test.csv')\nprint(f'test.csv : shape={test_df.shape}')","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:03:44.743714Z","iopub.execute_input":"2023-03-04T01:03:44.743981Z","iopub.status.idle":"2023-03-04T01:03:44.781965Z","shell.execute_reply.started":"2023-03-04T01:03:44.743956Z","shell.execute_reply":"2023-03-04T01:03:44.780609Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"test_df.session_id.unique()","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:03:44.783758Z","iopub.execute_input":"2023-03-04T01:03:44.784050Z","iopub.status.idle":"2023-03-04T01:03:44.796300Z","shell.execute_reply.started":"2023-03-04T01:03:44.784024Z","shell.execute_reply":"2023-03-04T01:03:44.795057Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## sample_submission.csv\n\n- 先ほどのテストデータの三人の回答\n- level_group列に各質問のレベルグループが記載されている\n- ここから、questionとlevelは下のように対応していることが分かる\n\n| question | 1-3 | 4-13 | 14-18 |\n| :---: | :---: | :---: | :---: |\n| level group | 0-4 | 5-12 | 13-22 |","metadata":{}},{"cell_type":"code","source":"%%time\nsample_submission_df = pd.read_csv('/kaggle/input/predict-student-performance-from-game-play/sample_submission.csv')\nprint(f'sample_submission.csv : shape={sample_submission_df.shape}')","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:03:44.797642Z","iopub.execute_input":"2023-03-04T01:03:44.797908Z","iopub.status.idle":"2023-03-04T01:03:44.820345Z","shell.execute_reply.started":"2023-03-04T01:03:44.797884Z","shell.execute_reply":"2023-03-04T01:03:44.819504Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"try:\n    get_session_id = lambda sq: sq.split('_')[0] \n    get_question_id = lambda sq: sq.split('_')[1] \n    get_level_group = lambda sq: sq.split('_')[1] \n\n    df_question_level = pd.DataFrame()\n    df_question_level['sesson_id'] = sample_submission_df.session_id.apply(get_session_id)\n    df_question_level['question_id'] = sample_submission_df.session_id.apply(get_question_id)\n    df_question_level['level_group'] = sample_submission_df.session_level.apply(get_level_group)\n    df_question_level['question_level'] = df_question_level['question_id'] + ':' + df_question_level['level_group']\n    for s in df_question_level.question_level.unique():\n        print(s)\nexcept:\n    pass","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:03:44.822881Z","iopub.execute_input":"2023-03-04T01:03:44.823337Z","iopub.status.idle":"2023-03-04T01:03:44.834769Z","shell.execute_reply.started":"2023-03-04T01:03:44.823305Z","shell.execute_reply":"2023-03-04T01:03:44.833952Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"---\n\n# BASELINE\n\n","metadata":{}},{"cell_type":"markdown","source":"### M. CHRIS DEOTTE\n\n- DEOTTE氏のBASELINEでは前処理として下記を実施している\n  - group_by(session_id, level_group)\n  - type==str : n_unique\n  - type==num : mean, std\n  - event_name : One-Hot-Encording\n  - elasped_time : sum\n- [XGBoost Baseline - [0.676]](https://www.kaggle.com/code/cdeotte/xgboost-baseline-0-676)  by CHRIS DEOTTE\n- [Predict Train Mean Baseline - [0.648]](https://www.kaggle.com/code/cdeotte/predict-train-mean-baseline-0-648) by CHRIS DEOTTE\n- [Random Forest Baseline - [0.664]](https://www.kaggle.com/code/cdeotte/random-forest-baseline-0-664) by CHRIS DEOTTE\n\n### M. NADARE\n\n- NADARE氏のBASELINEでは前処理として下記を実施している\n  - 使用する変数を限定 : `[\"event_name\", \"name\", \"level\", \"fqid\", \"room_fqid\", \"text_fqid\", \"text\"]`\n  - すべてをintに変換\n- [TensorFlow Baseline[LB 0.681]](https://www.kaggle.com/code/nadare/tensorflow-baseline-lb-0-681) by NADARE\n  - Deep Cross Network v2 を使用","metadata":{}},{"cell_type":"markdown","source":"---\n\n# 提出データ\n\n- `jo_wilder`ライブラリを使用し回答データを作成する\n- **注意:このライブラリを再実行したい場合はカーネルをリスタートする必要がある**\n- 提出方法の参考:[Basic Submission Demo](https://www.kaggle.com/code/philculliton/basic-submission-demo) by PHIL CULLITON\n  - score:0.226, ans=0で提出した例","metadata":{}},{"cell_type":"markdown","source":"### 予測モデルの作成\n\n- ここではサンプルという事で、運任せで回答を決定する。","metadata":{}},{"cell_type":"code","source":"def my_predict(session_id_, question_id_):\n    return np.random.randint(0,1)","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:03:44.836131Z","iopub.execute_input":"2023-03-04T01:03:44.836725Z","iopub.status.idle":"2023-03-04T01:03:44.844619Z","shell.execute_reply.started":"2023-03-04T01:03:44.836697Z","shell.execute_reply":"2023-03-04T01:03:44.843130Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- 配布ライブラリのインポート","metadata":{}},{"cell_type":"code","source":"import jo_wilder\nenv = jo_wilder.make_env()\niter_test = env.iter_test()","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:03:44.845910Z","iopub.execute_input":"2023-03-04T01:03:44.846583Z","iopub.status.idle":"2023-03-04T01:03:44.913722Z","shell.execute_reply.started":"2023-03-04T01:03:44.846533Z","shell.execute_reply":"2023-03-04T01:03:44.911726Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- jo_wilderのiteratorで各データを生成\n- env.predictに各回答を返すとsubmission.csvが生成される","metadata":{}},{"cell_type":"code","source":"np.random.seed(seed=46)\n\nget_session_id = lambda sq: sq.split('_')[0] \nget_question_id = lambda sq: sq.split('_')[1] \n\n# 用意されているiteratorでループする\nfor (sample_submission, test) in iter_test:\n    for index, row in sample_submission.iterrows():\n        sx = get_question_id(row.session_id)\n        qx = get_question_id(row.session_id)\n        sample_submission.loc[index,'correct'] = my_predict(sx, qx)\n        \n    # 用意されているpredict関数に自分の回答を提出\n    env.predict(sample_submission)","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:03:44.916448Z","iopub.execute_input":"2023-03-04T01:03:44.916779Z","iopub.status.idle":"2023-03-04T01:03:44.979064Z","shell.execute_reply.started":"2023-03-04T01:03:44.916750Z","shell.execute_reply":"2023-03-04T01:03:44.978257Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"**※ 提出先では別のデータセットが使用されるので注意**\n\n- 生成された提出ファイルを確認する\n- 1 or 0 ではなく 'True' or 'False' になっていないかよく確認する","metadata":{}},{"cell_type":"code","source":"df = pd.read_csv('submission.csv')\nprint('Sample submission shape:', df.shape )\nprint('Sample submission average prediction:', df.correct.mean() )\ndf.head()","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:03:44.980065Z","iopub.execute_input":"2023-03-04T01:03:44.980935Z","iopub.status.idle":"2023-03-04T01:03:44.995054Z","shell.execute_reply.started":"2023-03-04T01:03:44.980873Z","shell.execute_reply":"2023-03-04T01:03:44.993792Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"print(Path('submission.csv').open('r').read()[:100])","metadata":{"execution":{"iopub.status.busy":"2023-03-04T01:03:44.996368Z","iopub.execute_input":"2023-03-04T01:03:44.996736Z","iopub.status.idle":"2023-03-04T01:03:45.002815Z","shell.execute_reply.started":"2023-03-04T01:03:44.996710Z","shell.execute_reply":"2023-03-04T01:03:45.001954Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"- フォーマットなどを確認したらsubmitを実行する\n- この時、internet無効オプションを設定する必要がある\n- 提出先でNotebookが実行され、生成されたsubmission.csvが回答に用いられる\n- このノートの実行時間が9時間を超えないように設計する事\n\n![image.png](attachment:cc8af80b-2ea1-48b4-8985-92c7c503d535.png)\n","metadata":{},"attachments":{"cc8af80b-2ea1-48b4-8985-92c7c503d535.png":{"image/png":"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"}}},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}