{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"_kg_hide-input":true,"_kg_hide-output":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\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 read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_kg_hide-input":true},"cell_type":"code","source":"import plotly.express as px\nimport plotly.graph_objects as go\nimport plotly.figure_factory as ff\nfrom plotly.subplots import make_subplots\nimport matplotlib.pyplot as plt\n\nfrom pandas_profiling import ProfileReport\nimport seaborn as sns\nfrom sklearn import metrics\nfrom scipy import stats\n\nfrom copy import deepcopy\n\nfrom sklearn.ensemble import RandomForestRegressor\nfrom xgboost import XGBRegressor\n\nfrom sklearn.model_selection import GridSearchCV\nfrom sklearn.model_selection import KFold\nfrom sklearn.model_selection import cross_val_score\n\nimport optuna\nfrom optuna import Trial, visualization\n\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import f1_score,confusion_matrix\nfrom sklearn.metrics import accuracy_score, mean_squared_error","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Tabular Playground Series 📚 - Jan 2021 📈"},{"metadata":{},"cell_type":"markdown","source":"![tps-jan21.jpg](attachment:tps-jan21.jpg)","attachments":{"tps-jan21.jpg":{"image/jpeg":"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"}}},{"metadata":{"trusted":true},"cell_type":"code","source":"train_df = pd.read_csv('/kaggle/input/tabular-playground-series-jan-2021/train.csv')\ntest_df = pd.read_csv('/kaggle/input/tabular-playground-series-jan-2021/test.csv')\nsub_df = pd.read_csv('/kaggle/input/tabular-playground-series-jan-2021/sample_submission.csv')\n\ntrain_df.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"feature_cols = train_df.drop(['id', 'target'], axis=1).columns\n\nx = train_df[feature_cols]\ny = train_df['target']\n\nprint(x.shape, y.shape)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"## Join train and test datasets in order to obtain the same number of features during categorical conversion\ntrain_indexs = train_df.index\ntest_indexs = test_df.index\n\ndf =  pd.concat(objs=[train_df, test_df], axis=0).reset_index(drop=True)\ndf = df.drop('id', axis=1)\n\nlen(train_indexs), len(test_indexs)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 1. Data Visualization 📊"},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"def fix_skew(features):\n    \"\"\"\n    This function takes in a dataframe and return fixed skewed dataframe\n    \"\"\"\n    ## Import necessary modules \n    from scipy.special import boxcox1p\n    from scipy.stats import boxcox_normmax\n    \n    ## Getting all the data that are not of \"object\" type. \n    numerical_columns = features.select_dtypes(include=['int64','float64']).columns\n\n    # Check the skew of all numerical features\n    skewed_features = features[numerical_columns].apply(lambda x: stats.skew(x)).sort_values(ascending=False)\n    high_skew = skewed_features[abs(skewed_features) > 0.5]\n    skewed_features = high_skew.index\n\n    # Perform the Box-Cox transformation\n    for column in skewed_features:\n        features[column] = boxcox1p(features[column], boxcox_normmax(features[column] + 1))\n        \n    return features","execution_count":null,"outputs":[]},{"metadata":{"_kg_hide-input":true,"trusted":true},"cell_type":"code","source":"# I want to thanks @masumrumi for sharing this amazing plot!\ndef plotting_3_chart(df, feature):\n    ## Importing seaborn, matplotlab and scipy modules. \n    import seaborn as sns\n    import matplotlib.pyplot as plt\n    import matplotlib.gridspec as gridspec\n    from scipy import stats\n    import matplotlib.style as style\n    style.use('fivethirtyeight')\n\n    ## Creating a customized chart. and giving in figsize and everything. \n    fig = plt.figure(constrained_layout=True, figsize=(12,8))\n    ## creating a grid of 3 cols and 3 rows. \n    grid = gridspec.GridSpec(ncols=3, nrows=3, figure=fig)\n    #gs = fig3.add_gridspec(3, 3)\n\n    ## Customizing the histogram grid. \n    ax1 = fig.add_subplot(grid[0, :2])\n    ## Set the title. \n    ax1.set_title('Histogram')\n    ## plot the histogram. \n    sns.distplot(df.loc[:,feature], norm_hist=True, ax = ax1)\n\n    # customizing the QQ_plot. \n    ax2 = fig.add_subplot(grid[1, :2])\n    ## Set the title. \n    ax2.set_title('QQ_plot')\n    ## Plotting the QQ_Plot. \n    stats.probplot(df.loc[:,feature], plot = ax2)\n\n    ## Customizing the Box Plot. \n    ax3 = fig.add_subplot(grid[:, 2])\n    ## Set title. \n    ax3.set_title('Box Plot')\n    ## Plotting the box plot. \n    sns.boxplot(df.loc[:,feature], orient='v', ax = ax3 );","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df.info()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Target distribution"},{"metadata":{"trusted":true},"cell_type":"code","source":"plotting_3_chart(df, 'target')","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"As we can see, the target is close to binomial without much skewness. We could latter try to adjust it to a Gaussian."},{"metadata":{},"cell_type":"markdown","source":"## Features distribution analysis"},{"metadata":{"trusted":true},"cell_type":"code","source":"num_rows, num_cols = 4,4\n\nf, axes = plt.subplots(nrows=4, ncols=4, figsize=(12, 12))\n#f.suptitle('Distribution of Features', fontsize=16)\n\nfor index, column in enumerate(df[feature_cols].columns):\n    i,j = (index // num_cols, index % num_cols)\n    g = sns.distplot(train_df[column], color=\"m\", label=\"%.2f\"%(train_df[column].skew()), ax=axes[i,j])\n    g = g.legend(loc=\"best\")\n\n\nplt.tight_layout()\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Correlation analysis"},{"metadata":{"trusted":true},"cell_type":"code","source":"corr = df[feature_cols].corr().abs()\nmask = np.triu(np.ones_like(corr, dtype=np.bool))\n\nfig, ax = plt.subplots(figsize=(14, 14))\n\n# plot heatmap\nsns.heatmap(corr, mask=mask, annot=True, fmt=\".2f\", cmap='coolwarm',\n            cbar_kws={\"shrink\": .8})\n# yticks\nplt.yticks(rotation=0)\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"We can check that the variables are low correlated so we cna go ahead with the full set."},{"metadata":{},"cell_type":"markdown","source":"# 2. Feature Engineering ⚙️\n\nI am not going to do any feature engineering since the variables seems to be very clean and clear and there are no missing values.\n\nIn the future we can try to:\n\n* Try to transform target feature to better fit a normal distribution. \n* Sometime (not always) is useful to try a similar transformation for the model features as well; in their case, also scaling will improve the result with some of the models;"},{"metadata":{},"cell_type":"markdown","source":"# 3. Simple model: XGBoost Regressor"},{"metadata":{"trusted":true},"cell_type":"code","source":"param_grid = {\n    'n_estimators': [5, 10, 15, 20],\n    'max_depth': [2, 5, 7, 9]\n}\n\nx_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.3, random_state=42)\n\n# Instantiate model with 100 decision trees\nclf = XGBRegressor(random_state = 42)\n\nclf.fit(x_train, y_train)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Use the forest's predict method on the test data\npredictions = clf.predict(x_test)\n\n# Calculate the absolute errors\nerrors = abs(predictions - y_test)\n\n# Print out the mean absolute error (mae)\nprint('Mean Absolute Error:', round(np.mean(errors), 2), 'degrees.')\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 4. XGBoost Optuna Optimization"},{"metadata":{},"cell_type":"markdown","source":"![optuna-logo.png](attachment:optuna-logo.png)","attachments":{"optuna-logo.png":{"image/png":"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"}}},{"metadata":{"trusted":true},"cell_type":"code","source":"def objective(trial,data=x,target=y):\n    \n    train_x, test_x, train_y, test_y = train_test_split(data, target, test_size=0.15,random_state=42)\n    \n    # To select which parameters to optimize, please look at the XGBoost documentation:\n    # https://xgboost.readthedocs.io/en/latest/parameter.html\n    param = {\n        'tree_method':'gpu_hist',  # Use GPU acceleration\n        'lambda': trial.suggest_loguniform(\n            'lambda', 1e-3, 10.0\n        ),\n        'alpha': trial.suggest_loguniform(\n            'alpha', 1e-3, 10.0\n        ),\n        'colsample_bytree': trial.suggest_categorical(\n            'colsample_bytree', [0.5,0.6,0.7,0.8,0.9,1.0]\n        ),\n        'subsample': trial.suggest_categorical(\n            'subsample', [0.6,0.7,0.8,1.0]\n        ),\n        'learning_rate': trial.suggest_categorical(\n            'learning_rate', [0.008,0.009,0.01,0.012,0.014,0.016,0.018, 0.02]\n        ),\n        'n_estimators': trial.suggest_categorical(\n            \"n_estimators\", [150, 200, 300, 3000]\n        ),\n        'max_depth': trial.suggest_categorical(\n            'max_depth', [4,5,7,9,11,13,15,17]\n        ),\n        'random_state': 42,\n        'min_child_weight': trial.suggest_int(\n            'min_child_weight', 1, 300\n        ),\n    }\n    model = XGBRegressor(**param)  \n    \n    model.fit(train_x,train_y,eval_set=[(test_x,test_y)],early_stopping_rounds=100,verbose=False)\n    \n    preds = model.predict(test_x)\n    \n    rmse = mean_squared_error(test_y, preds,squared=False)\n    \n    return rmse","execution_count":null,"outputs":[]},{"metadata":{"trusted":true,"_kg_hide-output":true},"cell_type":"code","source":"study = optuna.create_study(direction='minimize')\nstudy.optimize(objective, n_trials=5)\nprint('Number of finished trials:', len(study.trials))\nprint('Best trial:', study.best_trial.params)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"study.trials_dataframe().head()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Visualization\n\nNow, that we have the optimization done, we can take a look of the output of the algorithm."},{"metadata":{"trusted":true},"cell_type":"code","source":"# plot_optimization_histor: shows the scores from all trials as well as the best score so far at each point.\noptuna.visualization.plot_optimization_history(study)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# plot_parallel_coordinate: interactively visualizes the hyperparameters and scores\noptuna.visualization.plot_parallel_coordinate(study)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# plot feature importance for algorithm parameters\nvisualization.plot_param_importances(study)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# finally plot best parameters\nstudy.best_params","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Visualize empirical distribution function\noptuna.visualization.plot_edf(study)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 5. Train final model\n\nNow we are going to train the final model with the best parameters"},{"metadata":{"trusted":true},"cell_type":"code","source":"best_params = study.best_params\nbest_params['tree_method'] = 'gpu_hist'\nbest_params['random_state'] = 42\n\nclf = XGBRegressor(**(best_params))\n\nclf.fit(x, y)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# 6. Submission"},{"metadata":{"trusted":true},"cell_type":"code","source":"preds = pd.Series(clf.predict(test_df.drop('id', axis=1)), name='target')\npreds = pd.concat([test_df['id'], preds], axis=1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"preds.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"preds.to_csv(\"submission.csv\", index=False)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"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":4,"nbformat_minor":4}