{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load in \n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the \"../input/\" directory.\n# For example, running this (by clicking run or pressing Shift+Enter) will list the files in the input directory\n\nimport os\nprint(os.listdir(\"../input\"))\n\n# Any results you write to the current directory are saved as output.","execution_count":null,"outputs":[]},{"metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","trusted":true},"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import mean_squared_error, cohen_kappa_score, accuracy_score\nfrom h2o.automl import H2OAutoML\nimport h2o\nh2o.init()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"3f47581ff3259016cb77cc88502435d2a68191f9"},"cell_type":"code","source":"%%time\nimport os\nimport json\nfrom tqdm import tqdm_notebook as tqdm\n\ndef load_desc_sentiment(path):\n    all_desc_sentiment_files = os.listdir(path)\n    count_file = len(all_desc_sentiment_files)\n    desc_sentiment_df = pd.DataFrame(columns=['PetID','desc_senti_magnitude','desc_senti_score'])\n    current_file_index = 1\n    for filename in tqdm(all_desc_sentiment_files):\n        with open(path+filename, 'r') as f:\n            sentiment_json = json.load(f)\n            petID = filename.split('.')[0]\n            magnitude = sentiment_json['documentSentiment']['magnitude']\n            score = sentiment_json['documentSentiment']['score']\n            desc_sentiment_df = desc_sentiment_df.append({'PetID': petID, 'desc_sentiment_df':magnitude,'desc_senti_score':score}, \\\n                                                         ignore_index=True)\n    \n    desc_sentiment_df['desc_sent_mult'] = desc_sentiment_df['desc_sentiment_df'] * desc_sentiment_df['desc_senti_score']\n    return desc_sentiment_df\n\ndf_train = pd.read_csv(\"../input/train/train.csv\")\ndf_train_sent = load_desc_sentiment(\"../input/train_sentiment/\")\n\ndf_test = pd.read_csv(\"../input/test/test.csv\")\ndf_test_sent = load_desc_sentiment(\"../input/test_sentiment/\")","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"de02d84eb0af9d23d216c7df6e81aecd5916d544"},"cell_type":"code","source":"def build_total_df(df_data, df_desc_sent):\n    df_total = pd.merge(df_data, df_desc_sent, on='PetID', how='outer')\n    df_total['words_count'] = df_train['Description'].fillna('').apply(lambda x: len(x.split()))\n    df_total['desc_len'] = df_train['Description'].fillna('').apply(lambda x: len(x))\n    df_total.fillna(0, inplace=True)\n    return df_total\n\ndf_total_train = build_total_df(df_train, df_train_sent)\ndf_total_test = build_total_df(df_test, df_test_sent)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"cf4016cbfeed0680c6ad8050b1e4e9e1a7c3ff40"},"cell_type":"code","source":"%%time\ndef kappa(y_true, y_pred):\n    return cohen_kappa_score(y_true, y_pred, weights='quadratic')\n\ndef regression_h2o(X_train, X_test, y_name, project_name):\n    aml = H2OAutoML(max_runtime_secs=180, seed=1, project_name=project_name)\n    aml.train(y=y_name, training_frame=X_train, leaderboard_frame=X_test)\n    return aml\n\nnumber_cols = df_total_train.describe().columns\ndata = df_total_train[number_cols]\nX_train, X_test = train_test_split(data)\ndf_total_train = df_total_train.fillna(0)\n\nprint(\"use cols:\", number_cols)\nX_train = h2o.H2OFrame(X_train.fillna(0))\nX_test = h2o.H2OFrame(X_test.fillna(0))\ny_name = 'AdoptionSpeed'\n\nmodel = regression_h2o(X_train, X_test, y_name, \"automl_h2o\")\nprint(model.leaderboard.head())\ny_pred = model.predict(X_test).as_data_frame()['predict'].values\ny_test = X_test.as_data_frame()[y_name].values\n# 0.23752238131720194\n# 0.522885 mse\ny_pred_int = y_pred.round()\nprint(\"new kappa:\", kappa(y_test, y_pred_int), \"mse:\", mean_squared_error(y_test, y_pred))\n# print(\"kappa:\", quadratic_weighted_kappa(y_test, y_pred), \"mse:\", mean_squared_error(y_test, y_pred))\n# print(\"kappa int:\", quadratic_weighted_kappa(y_test, y_pred_int), \"mse:\", mean_squared_error(y_test, y_pred_int), \"acc:\", accuracy_score(y_test, y_pred_int))\nprint()\n\n# 'Type', 'Age', 'Breed1', 'Breed2', 'Gender', 'Color1', 'Color2',\n#        'Color3', 'MaturitySize', 'FurLength', 'Vaccinated', 'Dewormed',\n#        'Sterilized', 'Health', 'Quantity', 'Fee', 'State', 'VideoAmt',\n#        'PhotoAmt', 'AdoptionSpeed', 'desc_senti_magnitude', 'words_count',\n#        'desc_len'\n\n# kappa: 0.6080146031303728 mse: 0.5448913813566739\n# kappa int: 0.6080146031303728 mse: 0.8823686316351027 acc: 0.40037343291544414\n# kappa int: 0.6192112374874424 mse: 0.8570285409442518 acc: 0.40037343291544414","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"1945a2f6c70bf93cb56281b4d0de7f5cd9948a74"},"cell_type":"code","source":"number_cols","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"a56120acd979aa5d1bab5a4777f9baa3ecc2b415"},"cell_type":"code","source":"df_test = df_total_test\ndf_test.fillna(0, inplace=True)\n\nX_submission = df_test[list(set(number_cols) - {'AdoptionSpeed'})]\ny_submission =  model.predict(h2o.H2OFrame(X_submission)).as_data_frame()['predict'].values\ny_submission_int = [int(x) for x in y_submission]\ndf_sample_submisson = pd.read_csv(\"../input/test/sample_submission.csv\")\ndf_sample_submisson['AdoptionSpeed'] = y_submission_int\ndf_sample_submisson.to_csv(\"submission.csv\", index=False)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_uuid":"d2a1e480789b4cb4eaa11d440bca8983188b49ca"},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.6.6","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":1}