{"cells":[{"metadata":{},"cell_type":"markdown","source":"# TalkingData AdTracking Fraud Detection Challenge\n# Can you detect fraudulent click traffic for mobile app ads?\n# https://www.kaggle.com/c/talkingdata-adtracking-fraud-detection"},{"metadata":{},"cell_type":"markdown","source":"**This notebook is inspired by an exercise in the [Feature Engineering](https://www.kaggle.com/learn/feature-engineering) course**  \n**It is also inspired by David Patton's notebook at [this link ](https://www.kaggle.com/dcpatton/td-fraud-detector-nn)**  \n**You can reference the tutorial at [this link](https://www.kaggle.com/matleonard/baseline-model)**  \n**You can reference my notebook at [this link](https://www.kaggle.com/georgezoto/feature-engineering-baseline-model)**  \n\n---\n"},{"metadata":{"trusted":true},"cell_type":"markdown","source":"# Introduction\n\nIn the exercise, you will work with data from the TalkingData AdTracking competition.  The goal of the competition is to predict if a user will download an app after clicking through an ad. \n\n<center><a href=\"https://www.kaggle.com/c/talkingdata-adtracking-fraud-detection\"><img src=\"https://i.imgur.com/srKxEkD.png\" width=600px></a></center>\n\nFor this course you will use a small sample of the data, dropping 99% of negative records (where the app wasn't downloaded) to make the target more balanced.\n\nAfter building a baseline model, you'll be able to see how your feature engineering and selection efforts improve the model's performance."},{"metadata":{},"cell_type":"markdown","source":"## Notes on strategy from 1st place winners\nhttps://www.kaggle.com/c/talkingdata-adtracking-fraud-detection/discussion/56475\n![image.png](attachment:image.png)","attachments":{"image.png":{"image/png":"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"}}},{"metadata":{},"cell_type":"markdown","source":"## Baseline Model\n\nThe first thing you'll do is construct a baseline model. We'll begin by looking at the data."},{"metadata":{},"cell_type":"markdown","source":"Data fields  \nEach row of the training data contains a click record, with the following features.  \n\n- ip: ip address of click.\n- app: app id for marketing.\n- device: device type id of user mobile phone (e.g., iphone 6 plus, iphone 7, huawei mate 7, etc.)\n- os: os version id of user mobile phone\n- channel: channel id of mobile ad publisher\n- click_time: timestamp of click (UTC)\n- attributed_time: if user download the app for after clicking an ad, this is the time of the app download\n- is_attributed: the target that is to be predicted, indicating the app was downloaded  \n\nNote that ip, app, device, os, and channel are encoded.\n\nThe test data is similar, with the following differences:\n- click_id: reference for making predictions\n- is_attributed: not included"},{"metadata":{"trusted":true},"cell_type":"code","source":"import dask\nimport dask.dataframe as dd\nimport numpy as np\nimport pandas as pd\nimport os\nimport time\n\nimport matplotlib.pyplot as plt\nplt.rcParams[\"figure.figsize\"] = [16,9]\n\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## ⚠️ Your notebook tried to allocate more memory than is available. It has restarted. ⚠️"},{"metadata":{},"cell_type":"markdown","source":"## Use smaller storage dtypes"},{"metadata":{"trusted":true},"cell_type":"code","source":"dtypes = {\n        'ip'            : 'uint32',\n        'app'           : 'object',\n        'device'        : 'object',\n        'os'            : 'object',\n        'channel'       : 'object',\n        'click_time'    : 'object',\n        'is_attributed' : 'uint8',\n        }","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Sample data - (100000, 8)\n#click_data = pd.read_csv('../input/talkingdata-adtracking-fraud-detection/train_sample.csv', parse_dates=['click_time'])\n\n#Full data - No idea how large it is, this notebook can not handle its size in RAM\n\n#Read only first limit rows\n#limit = 20_000_000\n\n#Read only these columns - skip attributed_time \nusecols = ['ip', 'app', 'device', 'os', 'channel', 'click_time', 'is_attributed']","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Competition data"},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_data = dd.read_csv('../input/talkingdata-adtracking-fraud-detection/train.csv', \n                               dtype=dtypes,\n                               #nrows=limit, #not supported by `dd.read_csv`\n                               usecols=usecols, \n                               parse_dates=['click_time'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_data.info()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_data_postitive = competition_data[competition_data.is_attributed == 1] ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"start_time = time.time()\ntype(competition_data_postitive)\nprint(\"--- %s seconds ---\" % (time.time() - start_time))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"start_time = time.time()\ncompetition_data_postitive = competition_data_postitive.compute()\nprint(\"--- %s seconds ---\" % (time.time() - start_time))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"print(type(competition_data_postitive), competition_data_postitive.shape)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_data_postitive.sample(10)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_data_negative = competition_data[competition_data.is_attributed == 0] ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_data_negative = competition_data_negative.sample(frac=0.0025) #number","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"start_time = time.time()\ncompetition_data_negative = competition_data_negative.compute()\nprint(\"--- %s seconds ---\" % (time.time() - start_time))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_data_negative.sample(10)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_competition_data = pd.concat([competition_data_postitive, competition_data_negative])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_competition_data.is_attributed.value_counts()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_competition_data.info()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Competition submission step"},{"metadata":{"trusted":true},"cell_type":"code","source":"test_dtypes = {\n        'click_id'      : 'uint32',\n        'ip'            : 'uint32',\n        'app'           : 'object',\n        'device'        : 'object',\n        'os'            : 'object',\n        'channel'       : 'object',\n        'click_time'    : 'object'\n        }","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"usecols","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data = dd.read_csv('../input/talkingdata-adtracking-fraud-detection/test.csv', \n                               dtype=test_dtypes,\n                               #nrows=limit, #not supported by `dd.read_csv`\n                               #usecols=usecols, #no columns needed to be skipped here\n                               parse_dates=['click_time'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"start_time = time.time()\ncompetition_test_data = competition_test_data.compute()\nprint(\"--- %s seconds ---\" % (time.time() - start_time))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"print(competition_test_data.shape)\ncompetition_test_data.head()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Feature Engineering 1 Construct features from timestamps\n\nNotice that the `click_data` DataFrame has a `'click_time'` column with timestamp data.\n\nUse this column to create features for the coresponding day, hour, minute and second. \n\nStore these as new integer columns `day`, `hour`, `minute`, and `second` in a new DataFrame `clicks`."},{"metadata":{"trusted":true},"cell_type":"code","source":"# Add new columns for timestamp features day, hour, minute, and second\nclicks = train_competition_data.copy()\nclicks['day'] = clicks['click_time'].dt.day.astype('uint8')\n# Fill in the rest\nclicks['hour'] = clicks['click_time'].dt.hour.astype('uint8')\nclicks['minute'] = clicks['click_time'].dt.minute.astype('uint8')\nclicks['second'] = clicks['click_time'].dt.second.astype('uint8')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"clicks.head()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Competition submission step"},{"metadata":{"trusted":true},"cell_type":"code","source":"# Add new columns for timestamp features day, hour, minute, and second\ncompetition_test_data = competition_test_data.copy()\ncompetition_test_data['day'] = competition_test_data['click_time'].dt.day.astype('uint8')\n# Fill in the rest\ncompetition_test_data['hour'] = competition_test_data['click_time'].dt.hour.astype('uint8')\ncompetition_test_data['minute'] = competition_test_data['click_time'].dt.minute.astype('uint8')\ncompetition_test_data['second'] = competition_test_data['click_time'].dt.second.astype('uint8')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data.head()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Question ??? \n\nclass sklearn.preprocessing.LabelEncoder[source]\nEncode target labels with value between 0 and n_classes-1.\n\nThis transformer should be used to encode **target values**, i.e. y, and not the input X."},{"metadata":{},"cell_type":"markdown","source":"## Feature Engineering 2 Label Encoding\nFor each of the categorical features `['ip', 'app', 'device', 'os', 'channel']`, use scikit-learn's `LabelEncoder` to create new features in the `clicks` DataFrame. The new column names should be the original column name with `'_labels'` appended, like `ip_labels`."},{"metadata":{},"cell_type":"markdown","source":"## ⚠️ ValueError: y contains previously unseen labels: [0, 1, 2,... ⚠️"},{"metadata":{},"cell_type":"markdown","source":"## 😀 Not the best solution to ValueError: y contains previously unseen labels: [0, 1, 2,... 😀\n## unknown_value = -1\n## ⚠️ Make sure this is int (as other labels) or you will not be able to predict in the end ⚠️\n## https://stackoverflow.com/questions/21057621/sklearn-labelencoder-with-never-seen-before-values"},{"metadata":{},"cell_type":"markdown","source":"## Workaround #2 potential data leakage???\n## http://kagglesolutions.com/r/feature-engineering--label-encoding\n## To resolve this issue we will first concatenate X_train and X_test together and then perform label encoding. You can have everything in a loop for all of your categorical features\n\n```\nX_train = pd.DataFrame({'x1': np.random.random(5), 'x2': ['cat', 'cat', 'dog', 'cat', 'dog']})\nX_test = pd.DataFrame({'x1': np.random.random(5), 'x2': ['cat', 'cat', 'dog', 'rat', 'dog']})\n\ncategorical_features = ['x2']\n\n# make an encoder object\nencoder = LabelEncoder()\n\n# fit and transform feature x2\nfor col in categorical_features:\n    encoder.fit(pd.concat([X_train[col], X_test[col]], axis=0, sort=False))\n    X_train[col] = encoder.transform(X_train[col])\n    X_test[col] = encoder.transform(X_test[col])\n    \nprint(X_train.head(), '\\n')\nprint(X_test.head(), '\\n')\n```"},{"metadata":{"trusted":true},"cell_type":"code","source":"# Not the best solution to ValueError: y contains previously unseen labels: [0, 1, 2,...\nunknown_value = -1 #Make sure this is int (as other labels) or you will not be able to predict in the end ⚠️\n\nfrom sklearn import preprocessing\n\ncat_features = ['ip', 'app', 'device', 'os', 'channel']\n#cat_features = ['ip']\n\n#encoder = preprocessing.LabelEncoder() - Incorrect, we need a label encoder for each feature\n# Create new columns in clicks using preprocessing.LabelEncoder()\n\nfor feature in cat_features:\n    start_time = time.time()\n    print(feature)\n    \n    #New encoder for each feature\n    encoder = preprocessing.LabelEncoder()\n    \n    #Fit on all possible values of this feature\n    encoder.fit(clicks[feature])\n    \n    #Create LabelEncoder of input to output\n    le_dict = dict(zip(encoder.classes_, encoder.transform(encoder.classes_)))\n    \n    #Encode unseen values to the unknown_value label\n    encoded = clicks[feature].apply(lambda x: le_dict.get(x, unknown_value))\n    clicks[feature+'_labels'] = encoded\n    \n    #Competition submission\n    competition_encoded = competition_test_data[feature].apply(lambda x: le_dict.get(x, unknown_value))\n    #ValueError: y contains previously unseen labels: [0, 2, 3, 4, 5,\n    competition_test_data[feature+'_labels'] = competition_encoded\n    \n    print(\"--- %s seconds ---\" % (time.time() - start_time))    ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"clicks.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data.head(20)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## dask_ml.preprocessing.LabelEncoder\nhttps://ml.dask.org/modules/generated/dask_ml.preprocessing.LabelEncoder.html"},{"metadata":{"trusted":true},"cell_type":"code","source":"#ip\n#--- 96.38041996955872 seconds ---\ndask_ml_preprocessing = \"\"\"\nstart_time = time.time()\n\n# Not the best solution to ValueError: y contains previously unseen labels: [0, 1, 2,...\nunknown_value = -1 #Make sure this is int (as other labels) or you will not be able to predict in the end ⚠️\n\n#from sklearn import preprocessing\nfrom dask_ml import preprocessing #Dask preprocessing\n\n#cat_features = ['ip', 'app', 'device', 'os', 'channel']\ncat_features = ['ip']\n\n#encoder = preprocessing.LabelEncoder() - Incorrect, we need a label encoder for each feature\n# Create new columns in clicks using preprocessing.LabelEncoder()\n\nfor feature in cat_features:\n    print(feature)\n    \n    #New encoder for each feature\n    encoder = preprocessing.LabelEncoder(use_categorical=False) #Dask Specify use_categorical=False to recover the scikit-learn behavior.\n    \n    #Fit on all possible values of this feature\n    encoder.fit(clicks[feature])\n    \n    #Create LabelEncoder of input to output\n    le_dict = dict(zip(encoder.classes_, encoder.transform(encoder.classes_)))\n    \n    #Encode unseen values to the unknown_value label\n    encoded = clicks[feature].apply(lambda x: le_dict.get(x, unknown_value))\n    clicks[feature+'_labels'] = encoded\n    \n    #Competition submission\n    competition_encoded = competition_test_data[feature].apply(lambda x: le_dict.get(x, unknown_value))\n    #ValueError: y contains previously unseen labels: [0, 2, 3, 4, 5,\n    competition_test_data[feature+'_labels'] = competition_encoded\n    \nprint(\"--- %s seconds ---\" % (time.time() - start_time))    \n\"\"\"","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## How many unknown_value did we get in the test dataset?"},{"metadata":{"trusted":true},"cell_type":"code","source":"train_ip_labels_unknowns = sum(clicks['ip_labels'] == unknown_value)\ntrain_ip_labels_unknowns","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"compet_test_ip_labels_unknowns = sum(competition_test_data['ip_labels'] == unknown_value)\ncompet_test_ip_labels_unknowns","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"my_own_metrics={'clicks': clicks.shape[0], #min(limit, clicks.shape[0]),\n                'competition_test_data':competition_test_data.shape[0],\n                'train ip_labels unknowns': train_ip_labels_unknowns,\n                'compet_test ip_labels unknowns':compet_test_ip_labels_unknowns,\n                'compet_test ip_labels unknowns %': round(100*compet_test_ip_labels_unknowns/competition_test_data.shape[0],2)}\nmy_own_metrics","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"clicks.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data.shape","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Feature Enginering 3 Add interactions\nSee also: https://www.kaggle.com/georgezoto/talkingdata-adtracking-comp-feature-generation"},{"metadata":{"trusted":true},"cell_type":"code","source":"import itertools\n\ncat_features = ['ip', 'app', 'device', 'os', 'channel']\ninteractions = pd.DataFrame(index=clicks.index)\n\n# Iterate through each pair of features, combine them into interaction features\nfor interaction_feature_tuple in itertools.combinations(cat_features,2):\n    #New feature name as concatination of 2 categorical features\n    interaction_feature  = '_'.join(list(interaction_feature_tuple))\n    print(interaction_feature_tuple, interaction_feature)\n    \n    #New interaction as concatination of the values of each combination of cateforical features\n    interactions_values = clicks[interaction_feature_tuple[0]].astype(str) + '_' + clicks[interaction_feature_tuple[1]].astype(str)\n    \n    #New label encoder for each interaction_feature \n    label_enc = preprocessing.LabelEncoder()\n    #interactions = interactions.assign(interaction_feature=label_enc.fit_transform(interactions_values)) ??? uses the string interaction_feature as the column name ???\n    #interactions[interaction_feature] = label_enc.fit_transform(interactions_values)                     #??? index values and how do they relate to the full dataset clicks ???\n\n    #Fit on all possible values of this feature\n    label_enc.fit(interactions_values)\n    #Create LabelEncoder of input to output\n    le_dict = dict(zip(label_enc.classes_, label_enc.transform(label_enc.classes_)))\n    #Encode unseen values to the unknown_value label\n    encoded = interactions_values.apply(lambda x: le_dict.get(x, unknown_value))\n    clicks[interaction_feature] = encoded\n    \n    print('clicks.columns')\n    print(clicks.columns)\n\n    #Competition submission\n    # Apply encoding to the competition test dataset\n    comp_interactions_values = competition_test_data[interaction_feature_tuple[0]].astype(str) + '_' + competition_test_data[interaction_feature_tuple[1]].astype(str)\n    #competition_test_data[interaction_feature] = label_enc.transform(comp_interactions_values)  #??? ValueError: y contains previously unseen labels: '119901_9' ???\n    \n    competition_encoded = comp_interactions_values.apply(lambda x: le_dict.get(x, unknown_value))\n    competition_test_data[interaction_feature] = competition_encoded\n    print('competition_test_data.columns')\n    print(competition_test_data.columns)\n    ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"clicks.columns","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data.columns","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Feature Enginering 4 Numerical Features Number of events in the past 6 hours"},{"metadata":{"trusted":true},"cell_type":"code","source":"def count_past_events_6h(series):\n    new_series = pd.Series(index=series, data=series.index, name=\"count_6_hours\").sort_index()\n    #launched = pd.Series(data =ks.index, index=ks.launched, name=\"count_7_days\").sort_index()\n    #print(new_series.head())\n    count_6_hours = new_series.rolling('6h').count() - 1\n    return count_6_hours","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"series = clicks[:100]['click_time']\nseries","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"new_series = pd.Series(index=series, data=series.index, name=\"count_6_hours\").sort_index()\nnew_series","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"new_series.rolling('6h').count() - 1","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#clicks['count_past_events_6h'] = \nevents_6h = count_past_events_6h(clicks[:100]['click_time'])\nevents_6h","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plt.plot(events_6h);","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#events_6h.index = events_6h.values\nevents_6h","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"events_6h = events_6h.reindex(clicks[:100].index)\nevents_6h","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#competition_test_data['count_past_events_6h'] = competition_test_data['click_time'].apply(count_past_events_6h)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Feature Enginering 5 Numerical Features Time since last event"},{"metadata":{"trusted":true},"cell_type":"code","source":"def time_diff_since_last_event(series):\n    \"\"\"Returns a series with the time since the last timestamp in seconds.\"\"\"\n    time_since_last_event = series.diff().dt.total_seconds()\n    return time_since_last_event","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"clicks['time_diff_since_last_event'] = clicks['click_time'].apply(time_diff_since_last_event)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data['time_diff_since_last_event'] = competition_test_data['click_time'].apply(time_diff_since_last_event)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Feature Enginering 6 Numerical Features Number of previous app downloads"},{"metadata":{"trusted":true},"cell_type":"code","source":"def previous_attributions(series):\n    \"\"\"Returns a series with the number of times an app has been downloaded.\"\"\"\n    #print(series)\n    #print(series.expanding(min_periods=2).sum())\n    sums = series.expanding(min_periods=2).sum() - series\n    return sums","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"clicks['time_diff_since_last_event'] = clicks['click_time'].apply(time_diff_since_last_event)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Machine Learning Model"},{"metadata":{},"cell_type":"markdown","source":"## Train, validation, and test sets\nWith our baseline features ready, we need to split our data into training and validation sets. We should also hold out a test set to measure the final accuracy of the model.\n\n### 4) Train/test splits with time series data\nThis is time series data. Are there any special considerations when creating train/test splits for time series? If so, what are they?"},{"metadata":{},"cell_type":"markdown","source":"### Create train/validation/test splits\n\nHere we'll create training, validation, and test splits. First, `clicks` DataFrame is sorted in order of increasing time. The first 80% of the rows are the train set, the next 10% are the validation set, and the last 10% are the test set."},{"metadata":{"trusted":true},"cell_type":"code","source":"clicks.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"feature_cols = ['day', 'hour', 'minute', 'second', \n                'ip_labels', 'app_labels', 'device_labels',\n                'os_labels', 'channel_labels']\n\nvalid_fraction = 0.1\nclicks_srt = clicks.sort_values('click_time')\nvalid_rows = int(len(clicks_srt) * valid_fraction)\ntrain = clicks_srt[:-valid_rows * 2]\n# valid size == test size, last two sections of the data\nvalid = clicks_srt[-valid_rows * 2:-valid_rows]\ntest = clicks_srt[-valid_rows:]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"print(clicks.shape,'\\n',train.shape,'\\n',valid.shape,'\\n',test.shape)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Train with LightGBM\n\nNow we can create LightGBM dataset objects for each of the smaller datasets and train the baseline model."},{"metadata":{"trusted":true},"cell_type":"code","source":"import lightgbm as lgb\n\ndtrain = lgb.Dataset(train[feature_cols], label=train['is_attributed'])\ndvalid = lgb.Dataset(valid[feature_cols], label=valid['is_attributed'])\ndtest = lgb.Dataset(test[feature_cols], label=test['is_attributed'])\n\nparam = {'num_leaves': 64, 'objective': 'binary'}\nparam['metric'] = 'auc'\nnum_round = 1000\n#bst = lgb.train(param, dtrain, num_round, valid_sets=[dvalid], early_stopping_rounds=10)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#type(bst) #lightgbm.basic.Booster","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### ??? TypeError: booster must be dict or LGBMModel\n\n- booster (dict or LGBMModel) – Dictionary returned from lightgbm.train() or LGBMModel instance.\nhttps://lightgbm.readthedocs.io/en/latest/pythonapi/lightgbm.plot_metric.html"},{"metadata":{"trusted":true},"cell_type":"code","source":"#lgb.plot_metric(bst, metric=metrics.roc_auc_score, dataset_names=[dtrain, dvalid, dtest]) #??? TypeError: booster must be dict or LGBMModel\n#, ax=None, xlim=None, ylim=None, title='Metric during training', xlabel='Iterations', ylabel='auto', figsize=None, dpi=None, grid=True)[source]","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"- evals_result (dict or None, optional (default=None)) –\n\nThis dictionary used to store all evaluation results of all the items in valid_sets.\n\nhttps://lightgbm.readthedocs.io/en/latest/pythonapi/lightgbm.train.html"},{"metadata":{},"cell_type":"markdown","source":"- validation_metrics inspired by:\nhttps://github.com/Microsoft/LightGBM/blob/2e93cdab9eee02d4d7f5cb3b6b31128dec94e25e/examples/python-guide/plot_example.py"},{"metadata":{"trusted":true},"cell_type":"code","source":"#Record eval results for plotting\nvalidation_metrics = {}  \n\nbst = lgb.train(param, \n                dtrain, \n                num_round, \n                valid_sets=[dvalid],\n                valid_names='Baseline Model',\n                early_stopping_rounds=10,\n                evals_result=validation_metrics,\n                verbose_eval=10)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"validation_metrics","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Plot validation AUC during training"},{"metadata":{"trusted":true},"cell_type":"code","source":"ax = lgb.plot_metric(validation_metrics, metric='auc');\n#plt.show();","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## ML Explainability and taking a closer look at feature importance, individual trees\nInspired by: https://github.com/Microsoft/LightGBM/blob/2e93cdab9eee02d4d7f5cb3b6b31128dec94e25e/examples/python-guide/plot_example.py\n"},{"metadata":{"trusted":true},"cell_type":"code","source":"print('Plot feature importances...')\nax = lgb.plot_importance(bst, max_num_features=15)\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"bst.num_trees()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"tree_index = 0\nprint('Plot '+str(tree_index)+'th tree...')  # one tree use categorical feature to split\nax = lgb.plot_tree(bst, tree_index=tree_index, figsize=(64, 36), show_info=['split_gain'])\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Evaluate the model\nFinally, with the model trained, we evaluate its performance on the test set. "},{"metadata":{"trusted":true},"cell_type":"code","source":"from sklearn import metrics\n\nstart_time = time.time()\n\nypred = bst.predict(test[feature_cols])\nscore = metrics.roc_auc_score(test['is_attributed'], ypred)\nprint(f\"Test score: {score}\")\n\nprint(\"--- %s seconds ---\" % (time.time() - start_time))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"my_own_metrics['test score'] = score","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"my_own_metrics","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#plt.bar(my_own_metrics.keys(), my_own_metrics.values())","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"This will be our baseline score for the model. When we transform features, add new ones, or perform feature selection, we should be improving on this score. However, since this is the test set, we only want to look at it at the end of all our manipulations. At the very end of this course you'll look at the test score again to see if you improved on the baseline model.\n\n# Keep Going\nNow that you have a baseline model, you are ready to **[use categorical encoding techniques](https://www.kaggle.com/matleonard/categorical-encodings)** to improve it."},{"metadata":{},"cell_type":"markdown","source":"---\n\n\n\n\n*Have questions or comments? Visit the [Learn Discussion forum](https://www.kaggle.com/learn-forum/161443) to chat with other Learners.*"},{"metadata":{},"cell_type":"markdown","source":"# Submit test predictions to TalkingData AdTracking Fraud Detection Challenge competition using the ***limited*** train.csv records from this notebook"},{"metadata":{"trusted":true},"cell_type":"code","source":"feature_cols + ['click_id']","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data = competition_test_data[feature_cols + ['click_id']]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test[feature_cols].head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"start_time = time.time()\ncompetition_predictions = bst.predict(competition_test_data[feature_cols])\nprint(\"--- %s seconds ---\" % (time.time() - start_time))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"type(competition_predictions)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_predictions","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_predictions_df = pd.DataFrame({'click_id': competition_test_data['click_id'],\n                                           'is_attributed': competition_predictions})\ncompetition_predictions_df","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Takes a long time\n#competition_predictions_df['click_id'] = competition_test_data['click_id']\n#competition_predictions_df = competition_predictions_df[['click_id', 'is_attributed']]\n#competition_predictions_df","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#competition_predictions_df['is_attributed'].value_counts().sort_index()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"pd.cut(competition_predictions_df['is_attributed'], bins=10).value_counts().sort_index()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"pd.cut(competition_predictions_df['is_attributed'], bins=10).value_counts().sort_index().plot(kind='bar', rot=45);","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#competition_predictions_df['is_attributed'].value_counts().sort_index().plot(kind='bar');","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#sum(competition_predictions_df['is_attributed'] <= 0.5)/competition_predictions_df.shape[0]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#sum(competition_predictions_df['is_attributed'] > 0.5)/competition_predictions_df.shape[0]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_predictions_df.to_csv('submission.csv', index=False)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"my_own_metrics['private score'] = 0\nmy_own_metrics['public score'] = 0\nmy_own_metrics","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Submit csv to competition\n<center><a href=\"https://www.kaggle.com/c/talkingdata-adtracking-fraud-detection\"><img src=\"https://i.imgur.com/srKxEkD.png\" width=400px></a></center>"},{"metadata":{},"cell_type":"markdown","source":"# Deep Learning Model"},{"metadata":{},"cell_type":"markdown","source":"## Data structure"},{"metadata":{"trusted":true},"cell_type":"code","source":"clicks.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data.head()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Import TensorFlow libraries"},{"metadata":{"trusted":true},"cell_type":"code","source":"import tensorflow as tf\n\nfrom tensorflow import keras\nfrom tensorflow.keras import layers\n\nprint(tf.__version__)\n\nimport random\nseed = 51\ntf.random.set_seed(seed)\nrandom.seed(seed)\n\nimport seaborn as sns\n\nimport matplotlib.pyplot as plt\nplt.rcParams[\"figure.figsize\"] = [16,9]","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Check if the dataset contains unknown values"},{"metadata":{"trusted":true},"cell_type":"code","source":"clicks.isna().sum()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"clicks.dtypes","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Split the data"},{"metadata":{"trusted":true},"cell_type":"code","source":"from sklearn.model_selection import train_test_split\n\ntrain_dataset, test_dataset = train_test_split(clicks, stratify=clicks['is_attributed'], test_size=0.2, random_state=seed)\ntrain_dataset, validation_dataset = train_test_split(train_dataset, stratify=train_dataset['is_attributed'], test_size=0.2, random_state=seed)\n\nprint(train_dataset.shape, validation_dataset.shape, test_dataset.shape)\nprint(100*train_dataset.shape[0]/clicks.shape[0], 100*validation_dataset.shape[0]/clicks.shape[0], 100*test_dataset.shape[0]/clicks.shape[0])","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Inspect the data"},{"metadata":{"trusted":true},"cell_type":"code","source":"sns.pairplot(train_dataset[['day', 'hour', 'minute', 'second', 'ip_labels', 'app_labels', 'device_labels', 'os_labels', 'channel_labels', 'is_attributed']], height=3)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Split features from labels"},{"metadata":{"trusted":true},"cell_type":"code","source":"train_labels = train_dataset.pop('is_attributed')\nvalidation_labels = validation_dataset.pop('is_attributed')\ntest_labels = test_dataset.pop('is_attributed')","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Keep only numerical features after label encoding above"},{"metadata":{"trusted":true},"cell_type":"code","source":"feature_col = ['day', 'hour', 'minute', 'second', 'ip_labels', 'app_labels', 'device_labels', 'os_labels', 'channel_labels']\n\n#Another way\n#train_dataset.select_dtypes(exclude='object')\n\ntrain_dataset = train_dataset[feature_col]\nvalidation_dataset = validation_dataset[feature_col]\ntest_dataset = test_dataset[feature_col]","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Normalize the data ⚠️ WRONG ⚠️ You get good performance on training** and validation but very bad results on comp test data"},{"metadata":{"trusted":true},"cell_type":"code","source":"def norm_WRONG(df):\n    return (df - df.mean()) / df.std()\n\n#AUC is around 0.55 with no normalization\n#normed_train_data = train_dataset\n#normed_validation_data = validation_dataset\n#normed_test_data = test_dataset","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Normalize the data"},{"metadata":{"trusted":true},"cell_type":"code","source":"def norm(df):\n    return (df - train_dataset.mean()) / train_dataset.std()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"normed_train_data = norm(train_dataset)\nnormed_validation_data = norm(validation_dataset)\nnormed_test_data = norm(test_dataset)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_dataset.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"normed_train_data.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"train_dataset['day'].hist();","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"normed_train_data['day'].hist();","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Useful custom methods"},{"metadata":{"trusted":true},"cell_type":"code","source":"def compute_metrics(y_true, y_pred):\n    tp = tf.keras.metrics.TruePositives()\n    tp.update_state(y_true, y_pred)\n    tp = int(tp.result().numpy())\n    fp = tf.keras.metrics.FalsePositives()\n    fp.update_state(y_true, y_pred)\n    fp = int(fp.result().numpy())\n    tn = tf.keras.metrics.TrueNegatives()\n    tn.update_state(y_true, y_pred)\n    tn = int(tn.result().numpy())\n    fn = tf.keras.metrics.FalseNegatives()\n    fn.update_state(y_true, y_pred)\n    fn = int(fn.result().numpy())\n    return [tp, fn, fp, tn]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def print_metrics(metrics):\n    print('True Positives: ' + str(metrics[0]))\n    print('True Negatives: ' + str(metrics[3]))\n    print('False Positives: ' + str(metrics[2]))\n    print('False Negatives: ' + str(metrics[1]))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def plot_results(history):\n    history = history.history\n\n    fig, (ax1, ax2, ax3) = plt.subplots(3, 1, sharex='col', figsize=(20, 14))\n\n    ax1.plot(history['loss'], label='Train loss')\n    ax1.plot(history['val_loss'], label='Validation loss')\n    ax1.legend(loc='best')\n    ax1.set_title('Loss')\n\n    ax2.plot(history['auprc'], label='Train AUPRC')\n    ax2.plot(history['val_auprc'], label='Validation AUPRC')\n    ax2.legend(loc='best')\n    ax2.set_title('AUPRC')\n    \n    \n    ax3.plot(history['auroc'], label='Train AUROC')\n    ax3.plot(history['val_auroc'], label='Validation AUROC')\n    ax3.legend(loc='best')\n    ax3.set_title('AUROC')\n\n    plt.xlabel('Epochs')\n    sns.despine()\n    plt.show()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## The model\n## Build the model"},{"metadata":{"trusted":true},"cell_type":"code","source":"train_dataset.dtypes","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def build_model():\n    model = keras.Sequential([\n    layers.Dense(64, activation='relu', input_shape=[len(train_dataset.keys())]),\n    layers.Dense(64, activation='relu'),\n    layers.Dense(1, activation='sigmoid')\n    ])\n\n    optimizer = tf.keras.optimizers.RMSprop(0.001)\n\n    model.compile(loss='binary_crossentropy',\n                optimizer=optimizer,\n                metrics=[tf.keras.metrics.AUC(curve='ROC', name='auroc'), \n                          tf.keras.metrics.AUC(curve='PR', name='auprc')])\n    model.summary()\n\n    return model","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model = build_model()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Now try out the model. Take a batch of 10 examples from the training data and call model.predict on it."},{"metadata":{"trusted":true},"cell_type":"code","source":"example_batch = normed_train_data[:10]\nexample_result = model.predict(example_batch)\nexample_result","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Train the model"},{"metadata":{"trusted":true},"cell_type":"code","source":"checkpoint_filepath = 'talkingdata-adtracking-dask-ml-dl-v11.h5'\n\nmodel_checkpoint = tf.keras.callbacks.ModelCheckpoint(checkpoint_filepath, verbose=0, save_weights_only=True, \n                                                      monitor='val_auprc', mode='max', save_best_only=True)\n\nearly_stopping = tf.keras.callbacks.EarlyStopping(monitor='val_auprc', patience=10, verbose=1, mode='max')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"EPOCHS = 100\nBATCH_SIZE = 256","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"tf.keras.backend.clear_session()\n\nhistory_v11 = model.fit(\n  normed_train_data, train_labels,\n  epochs=EPOCHS, \n  validation_data = (normed_validation_data, validation_labels), \n  batch_size=BATCH_SIZE,\n  verbose=2, \n  callbacks=[model_checkpoint, early_stopping]\n  )","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plot_results(history_v11)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Predict on a small portion of the training set a.k.a the test set"},{"metadata":{"trusted":true},"cell_type":"code","source":"y_pred = model.predict(normed_test_data)\nauprc = tf.keras.metrics.AUC(curve='PR')\nauprc.update_state(test_labels, y_pred)\nTF_Model_AUPRC = auprc.result().numpy()\nTF_Model_AUPRC","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"TF_Model_metrics = compute_metrics(test_labels, y_pred)\nprint_metrics(TF_Model_metrics)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#EPOCHS = 100\n#BATCH_SIZE = 256\n#0.90584815\n#True Positives: 73230\n#True Negatives: 362160\n#False Positives: 6734\n#False Negatives: 18139\n\n#Model v12\n#TF_Model_AUPRC 0.9084018 \n#TF_Model_AUROC 0.9549123\n#True Positives: 70773\n#True Negatives: 363975\n#False Positives: 4919\n#False Negatives: 20596\n    \n#EPOCHS = 100\n#BATCH_SIZE = 128\n#0.89784527\n#True Positives: 71936\n#True Negatives: 362700\n#False Positives: 6194\n#False Negatives: 19433","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test_labels[:5]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_pred[:5]","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Model v12"},{"metadata":{"trusted":true},"cell_type":"code","source":"def build_model_v12():\n    tf.keras.backend.clear_session()\n\n    model = keras.Sequential([\n    layers.Dense(128, activation='relu', input_shape=[len(train_dataset.keys())]),\n    layers.BatchNormalization(),\n    layers.Dropout(0.33), #Remove dropout or lower %\n    layers.Dense(64, activation='relu'),\n    layers.BatchNormalization(),\n    layers.Dropout(0.33),\n    layers.Dense(32, activation='relu'),\n    layers.BatchNormalization(),\n    layers.Dropout(0.33),\n    layers.Dense(1, activation='sigmoid')\n    ])\n\n    model.compile(loss='binary_crossentropy',\n                optimizer=tf.keras.optimizers.Adam(),\n                metrics=[tf.keras.metrics.AUC(curve='ROC', name='auroc'), \n                          tf.keras.metrics.AUC(curve='PR', name='auprc')])\n\n    model.summary()\n\n    return model","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model = build_model_v12()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"checkpoint_filepath = 'talkingdata-adtracking-dask-ml-dl-v12.h5'\n\nmodel_checkpoint = tf.keras.callbacks.ModelCheckpoint(checkpoint_filepath, verbose=0, save_weights_only=True, \n                                                      monitor='val_auprc', mode='max', save_best_only=True)\n\nearly_stopping = tf.keras.callbacks.EarlyStopping(monitor='val_auprc', patience=10, verbose=1, mode='max')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"EPOCHS = 100\nBATCH_SIZE = 256","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"tf.keras.backend.clear_session()\n\nhistory_v12 = model.fit(\n    normed_train_data, train_labels,\n    epochs=EPOCHS, \n    validation_data = (normed_validation_data, validation_labels), \n    batch_size=BATCH_SIZE,\n    verbose=2, \n    callbacks=[model_checkpoint, early_stopping]\n  )","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plot_results(history_v12)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"y_pred = model.predict(normed_test_data)\n\nauprc = tf.keras.metrics.AUC(curve='PR')\nauprc.update_state(test_labels, y_pred)\nTF_Model_AUPRC = auprc.result().numpy()\n\nauroc = tf.keras.metrics.AUC(curve='ROC')\nauroc.update_state(test_labels, y_pred)\nTF_Model_AUROC = auroc.result().numpy()\n\nprint('TF_Model_AUPRC',TF_Model_AUPRC, '\\nTF_Model_AUROC', TF_Model_AUROC)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"TF_Model_metrics = compute_metrics(test_labels, y_pred)\nprint_metrics(TF_Model_metrics)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Competition data "},{"metadata":{"trusted":true},"cell_type":"code","source":"feature_cols = ['day', 'hour', 'minute', 'second', \n                'ip_labels', 'app_labels', 'device_labels',\n                'os_labels', 'channel_labels']","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"test_dataset.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"normed_test_data.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data[feature_cols].head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data[feature_cols].dtypes","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data[feature_cols].isna().sum()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"col = 'day'\nprint(competition_test_data[col].mean(), competition_test_data[col].std())","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_test_data['day'].hist();","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"norm_competition_test_data = norm(competition_test_data[feature_cols])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"norm_competition_test_data['click_id'] = competition_test_data['click_id']","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"norm_competition_test_data.head()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## ⚠️ Your notebook tried to allocate more memory than is available. It has restarted. ⚠️"},{"metadata":{},"cell_type":"markdown","source":"## ⚠️ Partition competition_test_data so it fits in memory ⚠️"},{"metadata":{"trusted":true},"cell_type":"code","source":"total = norm_competition_test_data.shape[0]\nh = int(total/3)\n\ncompetition_test_data_partition1 = norm_competition_test_data.iloc[0:h]\ncompetition_test_data_partition2 = norm_competition_test_data.iloc[h:2*h]\ncompetition_test_data_partition3 = norm_competition_test_data.iloc[2*h:]\n\nprint(competition_test_data.shape, \n      competition_test_data_partition1.shape, \n      competition_test_data_partition2.shape, \n      competition_test_data_partition3.shape)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"start_time = time.time()\ncompetition_predictions_1 = model.predict(competition_test_data_partition1[feature_cols])\nprint(\"--- %s seconds ---\" % (time.time() - start_time))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_predictions_1_df = pd.DataFrame(competition_predictions_1, columns=['is_attributed'])\npd.cut(competition_predictions_1_df['is_attributed'], bins=10).value_counts().sort_index()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"pd.cut(competition_predictions_1_df['is_attributed'], bins=10).value_counts().sort_index().plot(kind='bar', rot=45);","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"start_time = time.time()\ncompetition_predictions_2 = model.predict(competition_test_data_partition2[feature_cols])\nprint(\"--- %s seconds ---\" % (time.time() - start_time))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_predictions_2_df = pd.DataFrame(competition_predictions_2, columns=['is_attributed'])\npd.cut(competition_predictions_2_df['is_attributed'], bins=10).value_counts().sort_index()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"pd.cut(competition_predictions_2_df['is_attributed'], bins=10).value_counts().sort_index().plot(kind='bar', rot=45);","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"start_time = time.time()\ncompetition_predictions_3 = model.predict(competition_test_data_partition3[feature_cols])\nprint(\"--- %s seconds ---\" % (time.time() - start_time))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_predictions_3_df = pd.DataFrame(competition_predictions_3, columns=['is_attributed'])\npd.cut(competition_predictions_3_df['is_attributed'], bins=10).value_counts().sort_index()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"pd.cut(competition_predictions_3_df['is_attributed'], bins=10).value_counts().sort_index().plot(kind='bar', rot=45);","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"competition_predictions = np.append(competition_predictions_1, competition_predictions_2)\ncompetition_predictions = np.append(competition_predictions, competition_predictions_3)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"submission_df = pd.read_csv('/kaggle/input/talkingdata-adtracking-fraud-detection/sample_submission.csv')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"submission_df['is_attributed'] = competition_predictions","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"submission_df.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"pd.cut(submission_df['is_attributed'], bins=10).value_counts().sort_index()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"pd.cut(submission_df['is_attributed'], bins=10).value_counts().sort_index().plot(kind='bar', rot=45);","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"submission_df.to_csv('dl_v11_label_encoding_submission.csv', index=False)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Submit csv to competition\n<center><a href=\"https://www.kaggle.com/c/talkingdata-adtracking-fraud-detection\"><img src=\"https://i.imgur.com/srKxEkD.png\" width=400px></a></center>"}],"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}