{"cells":[{"metadata":{},"cell_type":"markdown","source":"# Earthquake Prediction\nIn this kernel I will explore the data and try to use some different models to predict the time of the \"earthquake\". This is my first kernel so any feedback is appreciated!"},{"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)\nimport matplotlib.pyplot as plt\nimport scipy.signal\nimport scipy.stats\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":1,"outputs":[{"output_type":"stream","text":"['lanl-lstm-dataset', 'LANL-Earthquake-Prediction']\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"# 1) Load the Data"},{"metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","trusted":true},"cell_type":"code","source":"%%time\ntrain = pd.read_csv('../input/LANL-Earthquake-Prediction/train.csv', dtype={'acoustic_data': np.int16, 'time_to_failure': np.float32})","execution_count":2,"outputs":[{"output_type":"stream","text":"CPU times: user 2min 19s, sys: 12.4 s, total: 2min 31s\nWall time: 2min 32s\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"train.head()","execution_count":3,"outputs":[{"output_type":"execute_result","execution_count":3,"data":{"text/plain":"   acoustic_data  time_to_failure\n0             12           1.4691\n1              6           1.4691\n2              8           1.4691\n3              5           1.4691\n4              8           1.4691","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>acoustic_data</th>\n      <th>time_to_failure</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>12</td>\n      <td>1.4691</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>6</td>\n      <td>1.4691</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>8</td>\n      <td>1.4691</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>5</td>\n      <td>1.4691</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>8</td>\n      <td>1.4691</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"metadata":{},"cell_type":"markdown","source":"# 2) Visualize the Data\nI like to start by looking at the data, which in this case is a pretty simple time series."},{"metadata":{"trusted":true},"cell_type":"code","source":"%matplotlib inline\n\ntrain_acoustic_data_small = train['acoustic_data'].values[::50]\ntrain_time_to_failure_small = train['time_to_failure'].values[::50]\n\nfig, ax1 = plt.subplots(figsize=(16, 8))\nplt.title(\"Trends of acoustic_data and time_to_failure. 2% of data (sampled)\")\nplt.plot(train_acoustic_data_small, color='b')\nax1.set_ylabel('acoustic_data', color='b')\nplt.legend(['acoustic_data'])\nax2 = ax1.twinx()\nplt.plot(train_time_to_failure_small, color='g')\nax2.set_ylabel('time_to_failure', color='g')\nplt.legend(['time_to_failure'], loc=(0.875, 0.9))\nplt.grid(False)\n\ndel train_acoustic_data_small\ndel train_time_to_failure_small","execution_count":4,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 1152x576 with 2 Axes>","image/png":"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\n"},"metadata":{}}]},{"metadata":{},"cell_type":"markdown","source":"# 3) Generate the features\nHere I will calculate some features from a moving window similar to https://doi.org/10.1002/2017GL074677\n\nFeatures: Mean, variance, skewness, kurtosis (normalized and not)\n\nBEWARE OF LARGE DATA! You can't load the entire dataset into memory at once (At least I couldn't on my laptop).\n\n<a href=\"http://tinypic.com?ref=vcut1j\" target=\"_blank\"><img src=\"http://i66.tinypic.com/vcut1j.jpg\" border=\"0\"></a>"},{"metadata":{"trusted":true},"cell_type":"code","source":"%%time\nfeat_mat = np.zeros((15000,23))\nfs = 4000000 # Hz\nwindow_time = 0.0375 # seconds\noffset = 0.01 # seconds\nwindow_size = int(window_time*fs)\n\nfor i in np.arange(0,feat_mat.shape[0]):\n    start = int(i*offset*fs)\n    stop = int(window_size+i*offset*fs)\n    seg = train.iloc[start:stop,0]\n\n    feat_mat[i,0] = np.mean(seg)\n    feat_mat[i,1] = np.var(seg)\n    feat_mat[i,2] = scipy.stats.skew(seg)\n    feat_mat[i,3] = scipy.stats.kurtosis(seg)\n    feat_mat[i,-1] = train.iloc[stop,1]\n    ","execution_count":5,"outputs":[{"output_type":"stream","text":"CPU times: user 1min 13s, sys: 8 ms, total: 1min 13s\nWall time: 1min 13s\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"Next, we will calculate percentiles and add the time the signal spends above those \"thresholds\" as a feature. This is intuitive because you can see from the data, as the earthquake comes closer, the signal goes through bouts of higher magnitude. That is, the time the signal spends in the 91st percentile, for example, should go up as time_to_failure goes down. In the paper cited above, they calculate features based on the time spent above a certain threshold (1st to 9th and 91st to 99th)."},{"metadata":{"trusted":true},"cell_type":"code","source":"lower_perc = np.percentile(train.iloc[:,0],np.arange(1,10))\nupper_perc = np.percentile(train.iloc[:,0],np.arange(91,100))\n\nfor i in np.arange(0,feat_mat.shape[0]):\n    start = int(i*offset*fs)\n    stop = int(window_size+i*offset*fs)\n    seg = train.iloc[start:stop,0]\n    \n    for j in np.arange(0,lower_perc.shape[0]):\n        perc = np.size(np.where(seg>lower_perc[j])[0])\n        feat_mat[i,4+j] = perc/seg.shape[0] \n        \n    for j in np.arange(0,upper_perc.shape[0]):\n        perc = np.size(np.where(seg>upper_perc[j])[0])\n        feat_mat[i,13+j] = perc/seg.shape[0] \n\n    ","execution_count":6,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df = pd.DataFrame(feat_mat,columns=['mean', 'var', 'skew', 'kurt','perc1','perc2','perc3','perc4','perc5','perc6','perc7','perc8','perc9','perc91','perc92','perc93','perc94','perc95','perc96','perc97','perc98','perc99','time_to_failure'],dtype=np.float64)","execution_count":7,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df.head()","execution_count":8,"outputs":[{"output_type":"execute_result","execution_count":8,"data":{"text/plain":"       mean        var       ...           perc99  time_to_failure\n0  4.884113  26.021110       ...         0.006933         1.430797\n1  4.792853  29.356730       ...         0.006920         1.420198\n2  4.730867  38.116754       ...         0.007887         1.409599\n3  4.672107  39.427766       ...         0.008553         1.399996\n4  4.719380  47.123032       ...         0.011180         1.389297\n\n[5 rows x 23 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>mean</th>\n      <th>var</th>\n      <th>skew</th>\n      <th>kurt</th>\n      <th>perc1</th>\n      <th>perc2</th>\n      <th>perc3</th>\n      <th>perc4</th>\n      <th>perc5</th>\n      <th>perc6</th>\n      <th>perc7</th>\n      <th>perc8</th>\n      <th>perc9</th>\n      <th>perc91</th>\n      <th>perc92</th>\n      <th>perc93</th>\n      <th>perc94</th>\n      <th>perc95</th>\n      <th>perc96</th>\n      <th>perc97</th>\n      <th>perc98</th>\n      <th>perc99</th>\n      <th>time_to_failure</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>4.884113</td>\n      <td>26.021110</td>\n      <td>-0.024061</td>\n      <td>33.661319</td>\n      <td>0.993160</td>\n      <td>0.983327</td>\n      <td>0.972660</td>\n      <td>0.962893</td>\n      <td>0.947013</td>\n      <td>0.922867</td>\n      <td>0.922867</td>\n      <td>0.922867</td>\n      <td>0.884593</td>\n      <td>0.105367</td>\n      <td>0.071327</td>\n      <td>0.071327</td>\n      <td>0.049573</td>\n      <td>0.049573</td>\n      <td>0.035567</td>\n      <td>0.026733</td>\n      <td>0.016673</td>\n      <td>0.006933</td>\n      <td>1.430797</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>4.792853</td>\n      <td>29.356730</td>\n      <td>0.396437</td>\n      <td>103.267636</td>\n      <td>0.992987</td>\n      <td>0.982847</td>\n      <td>0.971840</td>\n      <td>0.962220</td>\n      <td>0.946300</td>\n      <td>0.922193</td>\n      <td>0.922193</td>\n      <td>0.922193</td>\n      <td>0.883427</td>\n      <td>0.097907</td>\n      <td>0.065753</td>\n      <td>0.065753</td>\n      <td>0.045547</td>\n      <td>0.045547</td>\n      <td>0.032853</td>\n      <td>0.024880</td>\n      <td>0.015607</td>\n      <td>0.006920</td>\n      <td>1.420198</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>4.730867</td>\n      <td>38.116754</td>\n      <td>0.531294</td>\n      <td>116.757047</td>\n      <td>0.991600</td>\n      <td>0.980613</td>\n      <td>0.968740</td>\n      <td>0.957947</td>\n      <td>0.941373</td>\n      <td>0.916060</td>\n      <td>0.916060</td>\n      <td>0.916060</td>\n      <td>0.875953</td>\n      <td>0.099907</td>\n      <td>0.068220</td>\n      <td>0.068220</td>\n      <td>0.047907</td>\n      <td>0.047907</td>\n      <td>0.035013</td>\n      <td>0.026893</td>\n      <td>0.017120</td>\n      <td>0.007887</td>\n      <td>1.409599</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>4.672107</td>\n      <td>39.427766</td>\n      <td>0.522032</td>\n      <td>109.758838</td>\n      <td>0.990887</td>\n      <td>0.979173</td>\n      <td>0.966607</td>\n      <td>0.955413</td>\n      <td>0.938573</td>\n      <td>0.912353</td>\n      <td>0.912353</td>\n      <td>0.912353</td>\n      <td>0.871813</td>\n      <td>0.098873</td>\n      <td>0.068760</td>\n      <td>0.068760</td>\n      <td>0.048893</td>\n      <td>0.048893</td>\n      <td>0.036173</td>\n      <td>0.028147</td>\n      <td>0.018373</td>\n      <td>0.008553</td>\n      <td>1.399996</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>4.719380</td>\n      <td>47.123032</td>\n      <td>0.348305</td>\n      <td>88.241359</td>\n      <td>0.988220</td>\n      <td>0.974980</td>\n      <td>0.961660</td>\n      <td>0.950373</td>\n      <td>0.933527</td>\n      <td>0.907420</td>\n      <td>0.907420</td>\n      <td>0.907420</td>\n      <td>0.867973</td>\n      <td>0.106913</td>\n      <td>0.075467</td>\n      <td>0.075467</td>\n      <td>0.054980</td>\n      <td>0.054980</td>\n      <td>0.041420</td>\n      <td>0.032753</td>\n      <td>0.022153</td>\n      <td>0.011180</td>\n      <td>1.389297</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"metadata":{},"cell_type":"markdown","source":"# 4) Apply models"},{"metadata":{"trusted":true},"cell_type":"code","source":"from sklearn.ensemble import RandomForestRegressor, AdaBoostRegressor, BaggingRegressor, GradientBoostingRegressor\nfrom sklearn.metrics import r2_score, mean_absolute_error\nfrom sklearn.model_selection import train_test_split, KFold, cross_val_score\nfrom sklearn.preprocessing import scale\nfrom sklearn.linear_model import LinearRegression, LassoCV, LassoLarsCV\nfrom keras.models import Sequential\nfrom keras.layers import Dense, LSTM\nfrom keras.wrappers.scikit_learn import KerasRegressor\nfrom sklearn.tree import DecisionTreeRegressor\nfrom sklearn.neural_network import MLPRegressor","execution_count":9,"outputs":[{"output_type":"stream","text":"Using TensorFlow backend.\n","name":"stderr"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"dataset = df.values\n\nX_train, X_test, y_train, y_test = train_test_split(dataset[:,:-1],dataset[:,-1],test_size=0.2)","execution_count":10,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## 4a) Linear Regression"},{"metadata":{"trusted":true},"cell_type":"code","source":"lr = LinearRegression()\nlr.fit(X_train, y_train)\npred_lr = lr.predict(X_test)\n\nprint(\"MAE = \",mean_absolute_error(y_test,pred_lr))","execution_count":11,"outputs":[{"output_type":"stream","text":"MAE =  2.127561284485727\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"## 4b) Random Forest\n\nPrimer on Random Forest:\n\nhttps://towardsdatascience.com/random-forest-in-python-24d0893d51c0"},{"metadata":{"trusted":true},"cell_type":"code","source":"# Instantiate model with 1000 decision trees\nrf = RandomForestRegressor(n_estimators = 100, random_state = 42)\n\n# Train the model on training data\nrf.fit(X_train, y_train)\n\npred_rf = rf.predict(X_test)\nprint(\"MAE = \",mean_absolute_error(y_test,pred_rf))","execution_count":12,"outputs":[{"output_type":"stream","text":"MAE =  2.049052494604521\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"## 4c) Neural Network\nSee https://machinelearningmastery.com/regression-tutorial-keras-deep-learning-library-python/"},{"metadata":{"trusted":true},"cell_type":"code","source":"NN = MLPRegressor()\nNN.fit(scale(X_train),y_train)\n\npred_nn = NN.predict(scale(X_test))\n\nprint(\"MAE = \", mean_absolute_error(y_test,pred_nn))","execution_count":13,"outputs":[{"output_type":"stream","text":"MAE =  2.0775148783549473\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"## 4d) Ensemble methods"},{"metadata":{},"cell_type":"markdown","source":"### 4di) Bagging\nMake a bunch of models and give them to the BaggingRegressor to see if it improves the score."},{"metadata":{"trusted":true},"cell_type":"code","source":"rf = RandomForestRegressor(n_estimators = 100, random_state = 42)\nlcv = LassoCV()\nllcv = LassoLarsCV()\nlr = LinearRegression()\nNN = MLPRegressor()\n\n# Put as many of the models as you want in this list\nclf_array = [rf,NN,lcv,llcv,lr]","execution_count":14,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"for clf in clf_array:\n    clf.fit(scale(X_train),y_train)\n    pred = clf.predict(scale(X_test))\n    vanilla_scores = mean_absolute_error(y_test,pred)\n    \n    bagging_clf = BaggingRegressor(clf, max_samples=0.25, max_features=1.0, random_state=27)\n    bagging_clf.fit(scale(X_train),y_train)\n    pred_bag = bagging_clf.predict(scale(X_test))\n    bag_scores = mean_absolute_error(y_test,pred_bag)\n    \n    print(\"vanilla {}: {}\",clf,vanilla_scores)\n    print(\"bagging {}: {}\",clf,bag_scores)\n\n    ","execution_count":15,"outputs":[{"output_type":"stream","text":"vanilla {}: {} RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n           max_features='auto', max_leaf_nodes=None,\n           min_impurity_decrease=0.0, min_impurity_split=None,\n           min_samples_leaf=1, min_samples_split=2,\n           min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=None,\n           oob_score=False, random_state=42, verbose=0, warm_start=False) 2.054723610614454\nbagging {}: {} RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n           max_features='auto', max_leaf_nodes=None,\n           min_impurity_decrease=0.0, min_impurity_split=None,\n           min_samples_leaf=1, min_samples_split=2,\n           min_weight_fraction_leaf=0.0, n_estimators=100, n_jobs=None,\n           oob_score=False, random_state=42, verbose=0, warm_start=False) 2.077799701599944\n","name":"stdout"},{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:562: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n  % self.max_iter, ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:562: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n  % self.max_iter, ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:562: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n  % self.max_iter, ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:562: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n  % self.max_iter, ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:562: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n  % self.max_iter, ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:562: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n  % self.max_iter, ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:562: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n  % self.max_iter, ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/neural_network/multilayer_perceptron.py:562: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (200) reached and the optimization hasn't converged yet.\n  % self.max_iter, ConvergenceWarning)\n","name":"stderr"},{"output_type":"stream","text":"vanilla {}: {} MLPRegressor(activation='relu', alpha=0.0001, batch_size='auto', beta_1=0.9,\n       beta_2=0.999, early_stopping=False, epsilon=1e-08,\n       hidden_layer_sizes=(100,), learning_rate='constant',\n       learning_rate_init=0.001, max_iter=200, momentum=0.9,\n       n_iter_no_change=10, nesterovs_momentum=True, power_t=0.5,\n       random_state=None, shuffle=True, solver='adam', tol=0.0001,\n       validation_fraction=0.1, verbose=False, warm_start=False) 2.06480798905843\nbagging {}: {} MLPRegressor(activation='relu', alpha=0.0001, batch_size='auto', beta_1=0.9,\n       beta_2=0.999, early_stopping=False, epsilon=1e-08,\n       hidden_layer_sizes=(100,), learning_rate='constant',\n       learning_rate_init=0.001, max_iter=200, momentum=0.9,\n       n_iter_no_change=10, nesterovs_momentum=True, power_t=0.5,\n       random_state=None, shuffle=True, solver='adam', tol=0.0001,\n       validation_fraction=0.1, verbose=False, warm_start=False) 2.077890001743413\n","name":"stdout"},{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n","name":"stderr"},{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n","name":"stderr"},{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n","name":"stderr"},{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n","name":"stderr"},{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n","name":"stderr"},{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/coordinate_descent.py:492: ConvergenceWarning: Objective did not converge. You might want to increase the number of iterations. Fitting data with very small alpha may cause precision problems.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=8.255e-03, with an active set of 2 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 5 iterations, i.e. alpha=3.566e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 14 iterations, i.e. alpha=1.766e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 14 iterations, i.e. alpha=1.589e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 14 iterations, i.e. alpha=1.587e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 18 iterations, alpha=1.113e-03, previous alpha=6.833e-04, with an active set of 13 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=8.860e-03, with an active set of 2 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 5 iterations, i.e. alpha=3.571e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 14 iterations, i.e. alpha=1.710e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 14 iterations, i.e. alpha=1.646e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 14 iterations, i.e. alpha=1.579e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 17 iterations, i.e. alpha=6.584e-04, with an active set of 13 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 18 iterations, alpha=1.005e-03, previous alpha=6.158e-04, with an active set of 13 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 15 iterations, i.e. alpha=5.594e-05, with an active set of 7 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 16 iterations, i.e. alpha=5.541e-05, with an active set of 8 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 20 iterations, i.e. alpha=2.771e-05, with an active set of 10 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=1.329e-05, with an active set of 12 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=1.235e-05, with an active set of 12 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 24 iterations, i.e. alpha=8.519e-06, with an active set of 14 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 26 iterations, i.e. alpha=5.865e-06, with an active set of 16 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 26 iterations, i.e. alpha=2.140e-06, with an active set of 16 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 27 iterations, i.e. alpha=7.832e-07, with an active set of 17 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 27 iterations, i.e. alpha=7.832e-07, with an active set of 17 regressors, and the smallest cholesky pivot element being 1.825e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 28 iterations, i.e. alpha=2.856e-07, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.825e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 28 iterations, i.e. alpha=2.856e-07, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.332e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 7 iterations, i.e. alpha=4.910e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 9 iterations, i.e. alpha=3.883e-03, with an active set of 5 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 16 iterations, i.e. alpha=2.254e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 16 iterations, i.e. alpha=2.117e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 17 iterations, i.e. alpha=1.466e-03, with an active set of 11 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 18 iterations, alpha=2.319e-03, previous alpha=1.211e-03, with an active set of 11 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 30 iterations, i.e. alpha=5.194e-06, with an active set of 16 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 30 iterations, i.e. alpha=5.194e-06, with an active set of 16 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 31 iterations, i.e. alpha=2.020e-06, with an active set of 17 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 31 iterations, i.e. alpha=2.020e-06, with an active set of 17 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 31 iterations, i.e. alpha=2.020e-06, with an active set of 17 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 32 iterations, i.e. alpha=5.038e-07, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 32 iterations, i.e. alpha=5.038e-07, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 32 iterations, i.e. alpha=5.038e-07, with an active set of 18 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.602e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 5 iterations, i.e. alpha=8.012e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=2.189e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 5 iterations, i.e. alpha=9.151e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 13 iterations, alpha=5.571e-03, previous alpha=5.302e-03, with an active set of 8 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=2.331e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 5 iterations, i.e. alpha=9.899e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 16 iterations, i.e. alpha=4.699e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 16 iterations, i.e. alpha=4.685e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 16 iterations, i.e. alpha=4.349e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 17 iterations, i.e. alpha=1.898e-03, with an active set of 11 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 17 iterations, i.e. alpha=1.837e-03, with an active set of 11 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 19 iterations, alpha=2.217e-03, previous alpha=1.753e-03, with an active set of 12 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.592e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 7 iterations, i.e. alpha=5.713e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 10 iterations, alpha=8.791e-03, previous alpha=5.085e-03, with an active set of 5 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.667e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 27 iterations, i.e. alpha=5.272e-06, with an active set of 17 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n","name":"stderr"},{"output_type":"stream","text":"vanilla {}: {} LassoCV(alphas=None, copy_X=True, cv='warn', eps=0.001, fit_intercept=True,\n    max_iter=1000, n_alphas=100, n_jobs=None, normalize=False,\n    positive=False, precompute='auto', random_state=None,\n    selection='cyclic', tol=0.0001, verbose=False) 2.1331388947207053\nbagging {}: {} LassoCV(alphas=None, copy_X=True, cv='warn', eps=0.001, fit_intercept=True,\n    max_iter=1000, n_alphas=100, n_jobs=None, normalize=False,\n    positive=False, precompute='auto', random_state=None,\n    selection='cyclic', tol=0.0001, verbose=False) 2.131284373223639\n","name":"stdout"},{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 28 iterations, i.e. alpha=2.014e-06, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 28 iterations, i.e. alpha=2.014e-06, with an active set of 18 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=2.161e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 5 iterations, i.e. alpha=1.080e-02, with an active set of 3 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 15 iterations, i.e. alpha=4.781e-03, with an active set of 9 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 19 iterations, i.e. alpha=2.257e-03, with an active set of 13 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 20 iterations, i.e. alpha=1.499e-03, with an active set of 14 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=7.381e-04, with an active set of 16 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=6.112e-04, with an active set of 16 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=5.906e-04, with an active set of 16 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 24 iterations, i.e. alpha=1.897e-04, with an active set of 18 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 24 iterations, i.e. alpha=1.820e-04, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 24 iterations, i.e. alpha=1.820e-04, with an active set of 18 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.916e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 5 iterations, i.e. alpha=9.191e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 12 iterations, i.e. alpha=4.366e-03, with an active set of 8 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 13 iterations, i.e. alpha=3.877e-03, with an active set of 9 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 17 iterations, i.e. alpha=2.183e-03, with an active set of 11 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 20 iterations, i.e. alpha=9.544e-04, with an active set of 14 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 20 iterations, i.e. alpha=8.509e-04, with an active set of 14 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=7.026e-04, with an active set of 16 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 23 iterations, i.e. alpha=2.364e-04, with an active set of 17 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 24 iterations, i.e. alpha=1.486e-04, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 24 iterations, i.e. alpha=1.486e-04, with an active set of 18 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 24 iterations, i.e. alpha=2.343e-07, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.439e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 5 iterations, i.e. alpha=6.737e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 14 iterations, i.e. alpha=3.149e-03, with an active set of 8 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 14 iterations, i.e. alpha=2.674e-03, with an active set of 8 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 18 iterations, i.e. alpha=1.191e-03, with an active set of 12 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 19 iterations, i.e. alpha=1.123e-03, with an active set of 13 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 22 iterations, alpha=2.296e-03, previous alpha=6.288e-04, with an active set of 15 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.834e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 5 iterations, i.e. alpha=8.325e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 14 iterations, i.e. alpha=4.649e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 14 iterations, i.e. alpha=3.676e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 15 iterations, i.e. alpha=3.385e-03, with an active set of 11 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 16 iterations, i.e. alpha=2.059e-03, with an active set of 12 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 18 iterations, i.e. alpha=1.465e-03, with an active set of 14 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 19 iterations, alpha=3.169e-03, previous alpha=1.334e-03, with an active set of 14 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.979e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 5 iterations, i.e. alpha=9.089e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 14 iterations, i.e. alpha=5.304e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 15 iterations, i.e. alpha=3.741e-03, with an active set of 11 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 15 iterations, i.e. alpha=3.682e-03, with an active set of 11 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 15 iterations, i.e. alpha=3.330e-03, with an active set of 11 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 17 iterations, i.e. alpha=1.970e-03, with an active set of 13 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 18 iterations, alpha=3.192e-03, previous alpha=1.970e-03, with an active set of 13 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 16 iterations, i.e. alpha=1.107e-04, with an active set of 8 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 17 iterations, i.e. alpha=1.094e-04, with an active set of 9 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 19 iterations, i.e. alpha=4.974e-05, with an active set of 11 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=3.583e-05, with an active set of 14 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 23 iterations, i.e. alpha=2.804e-05, with an active set of 15 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 23 iterations, i.e. alpha=2.534e-05, with an active set of 15 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 24 iterations, i.e. alpha=1.668e-05, with an active set of 16 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 25 iterations, i.e. alpha=5.853e-06, with an active set of 17 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 25 iterations, i.e. alpha=2.686e-06, with an active set of 17 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 26 iterations, i.e. alpha=2.296e-06, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 26 iterations, i.e. alpha=2.255e-07, with an active set of 18 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 26 iterations, i.e. alpha=2.255e-07, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=1.736e-05, with an active set of 14 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=1.736e-05, with an active set of 14 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 24 iterations, alpha=1.384e-05, previous alpha=7.912e-06, with an active set of 15 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 15 iterations, i.e. alpha=1.053e-04, with an active set of 9 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 4 iterations, i.e. alpha=1.303e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 9 iterations, i.e. alpha=6.579e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 9 iterations, i.e. alpha=6.309e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 9 iterations, i.e. alpha=6.041e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 10 iterations, i.e. alpha=5.905e-03, with an active set of 4 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 23 iterations, i.e. alpha=3.819e-03, with an active set of 9 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 23 iterations, i.e. alpha=3.007e-03, with an active set of 9 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 25 iterations, i.e. alpha=2.377e-03, with an active set of 11 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 25 iterations, i.e. alpha=2.377e-03, with an active set of 11 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 26 iterations, alpha=3.485e-03, previous alpha=2.306e-03, with an active set of 11 regressors.\n  ConvergenceWarning)\n","name":"stderr"},{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.484e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 5 iterations, i.e. alpha=6.491e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 12 iterations, i.e. alpha=4.056e-03, with an active set of 8 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 14 iterations, alpha=3.173e-03, previous alpha=3.120e-03, with an active set of 9 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.141e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 7 iterations, i.e. alpha=4.667e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 9 iterations, i.e. alpha=3.754e-03, with an active set of 5 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 13 iterations, alpha=3.322e-03, previous alpha=2.746e-03, with an active set of 8 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=2.140e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 5 iterations, i.e. alpha=9.172e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 13 iterations, i.e. alpha=4.906e-03, with an active set of 7 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 13 iterations, i.e. alpha=4.585e-03, with an active set of 7 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 15 iterations, i.e. alpha=4.089e-03, with an active set of 9 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 20 iterations, i.e. alpha=2.044e-03, with an active set of 12 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=1.257e-03, with an active set of 14 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=1.257e-03, with an active set of 14 regressors, and the smallest cholesky pivot element being 2.107e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 23 iterations, i.e. alpha=1.003e-03, with an active set of 15 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 25 iterations, i.e. alpha=6.451e-04, with an active set of 17 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 25 iterations, i.e. alpha=5.598e-04, with an active set of 17 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 25 iterations, i.e. alpha=5.119e-04, with an active set of 17 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 25 iterations, i.e. alpha=3.081e-04, with an active set of 17 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 26 iterations, i.e. alpha=6.975e-05, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 26 iterations, i.e. alpha=6.975e-05, with an active set of 18 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.575e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 7 iterations, i.e. alpha=6.717e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 11 iterations, alpha=1.074e-02, previous alpha=5.920e-03, with an active set of 6 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.588e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 28 iterations, i.e. alpha=4.497e-07, with an active set of 18 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 28 iterations, i.e. alpha=4.497e-07, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.694e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 5 iterations, i.e. alpha=7.551e-03, with an active set of 3 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 11 iterations, i.e. alpha=4.173e-03, with an active set of 7 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 14 iterations, i.e. alpha=3.501e-03, with an active set of 10 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 15 iterations, i.e. alpha=2.942e-03, with an active set of 11 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 17 iterations, alpha=2.971e-03, previous alpha=1.876e-03, with an active set of 12 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 20 iterations, i.e. alpha=3.505e-05, with an active set of 8 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 30 iterations, i.e. alpha=1.230e-06, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 30 iterations, i.e. alpha=1.230e-06, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.825e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 30 iterations, i.e. alpha=1.230e-06, with an active set of 18 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.383e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/model_selection/_split.py:2053: FutureWarning: You should specify a value for 'cv' instead of relying on the default value. The default value will change from 3 to 5 in version 0.22.\n  warnings.warn(CV_WARNING, FutureWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=1.873e-05, with an active set of 10 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 23 iterations, i.e. alpha=1.809e-05, with an active set of 11 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 26 iterations, i.e. alpha=1.195e-05, with an active set of 14 regressors, and the smallest cholesky pivot element being 1.054e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 30 iterations, alpha=9.738e-06, previous alpha=8.646e-06, with an active set of 15 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 4 iterations, i.e. alpha=1.685e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 10 iterations, i.e. alpha=8.037e-03, with an active set of 4 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 13 iterations, i.e. alpha=7.158e-03, with an active set of 7 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 13 iterations, i.e. alpha=6.773e-03, with an active set of 7 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 13 iterations, i.e. alpha=6.599e-03, with an active set of 7 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 17 iterations, i.e. alpha=3.282e-03, with an active set of 11 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 18 iterations, i.e. alpha=2.885e-03, with an active set of 12 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 19 iterations, alpha=2.832e-03, previous alpha=2.725e-03, with an active set of 12 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 19 iterations, i.e. alpha=2.086e-05, with an active set of 9 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 21 iterations, i.e. alpha=1.720e-05, with an active set of 11 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=1.368e-05, with an active set of 12 regressors, and the smallest cholesky pivot element being 1.490e-08. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 22 iterations, i.e. alpha=1.368e-05, with an active set of 12 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:337: ConvergenceWarning: Early stopping the lars path, as the residues are small and the current value of alpha is no longer well controlled. 24 iterations, alpha=1.225e-05, previous alpha=6.806e-06, with an active set of 13 regressors.\n  ConvergenceWarning)\n/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/least_angle.py:311: ConvergenceWarning: Regressors in active set degenerate. Dropping a regressor, after 2 iterations, i.e. alpha=1.363e-02, with an active set of 2 regressors, and the smallest cholesky pivot element being 2.220e-16. Reduce max_iter or increase eps parameters.\n  ConvergenceWarning)\n","name":"stderr"},{"output_type":"stream","text":"vanilla {}: {} LassoLarsCV(copy_X=True, cv='warn', eps=2.220446049250313e-16,\n      fit_intercept=True, max_iter=500, max_n_alphas=1000, n_jobs=None,\n      normalize=True, positive=False, precompute='auto', verbose=False) 2.256867258583921\nbagging {}: {} LassoLarsCV(copy_X=True, cv='warn', eps=2.220446049250313e-16,\n      fit_intercept=True, max_iter=500, max_n_alphas=1000, n_jobs=None,\n      normalize=True, positive=False, precompute='auto', verbose=False) 3.194999983988803\nvanilla {}: {} LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None,\n         normalize=False) 2.1303775248618035\nbagging {}: {} LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None,\n         normalize=False) 2.131023100503804\n","name":"stdout"}]},{"metadata":{},"cell_type":"markdown","source":"# Load the test data and make predictions"},{"metadata":{"trusted":true},"cell_type":"code","source":"from tqdm import tqdm_notebook\n\nsubmission = pd.read_csv('../input/LANL-Earthquake-Prediction/sample_submission.csv', index_col='seg_id', dtype={\"time_to_failure\": np.float32})\n\nX_test = pd.DataFrame(columns=df.columns, dtype=np.float64, index=submission.index)\n\nfor i, seg_id in enumerate(tqdm_notebook(X_test.index)):\n    seg = pd.read_csv('../input/LANL-Earthquake-Prediction/test/' + seg_id + '.csv')\n    X_test.loc[seg_id, 'mean'] = np.mean(seg.values)\n    X_test.loc[seg_id, 'var'] = np.var(seg.values)\n    X_test.loc[seg_id, 'skew'] = scipy.stats.skew(seg.values)\n    X_test.loc[seg_id, 'kurt'] = scipy.stats.kurtosis(seg.values)\n    \n    for j in np.arange(0,9):\n        perc = np.size(np.where(seg>lower_perc[j])[0])\n        X_test.loc[seg_id, 'perc{}'.format(j+1)] = perc/seg.shape[0] \n        \n        perc = np.size(np.where(seg>upper_perc[j])[0])\n        X_test.loc[seg_id, 'perc9{}'.format(j+1)] = perc/seg.shape[0]\n\n    ","execution_count":16,"outputs":[{"output_type":"display_data","data":{"text/plain":"HBox(children=(IntProgress(value=0, max=2624), HTML(value='')))","application/vnd.jupyter.widget-view+json":{"version_major":2,"version_minor":0,"model_id":"c8feee3434a24897bd2532784ec89fff"}},"metadata":{}},{"output_type":"stream","text":"\n","name":"stdout"}]},{"metadata":{"trusted":true},"cell_type":"code","source":"X_test.drop(columns=['time_to_failure'],inplace=True)\nX_test.head()","execution_count":17,"outputs":[{"output_type":"execute_result","execution_count":17,"data":{"text/plain":"                mean        var    ...       perc98    perc99\nseg_id                             ...                       \nseg_00030f  4.491780  23.948039    ...     0.014080  0.006227\nseg_0012b5  4.171153  35.079793    ...     0.017787  0.009287\nseg_00184e  4.610260  48.260349    ...     0.017587  0.009667\nseg_003339  4.531473  16.926089    ...     0.007700  0.003720\nseg_0042cc  4.128340  33.606882    ...     0.015993  0.008233\n\n[5 rows x 22 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>mean</th>\n      <th>var</th>\n      <th>skew</th>\n      <th>kurt</th>\n      <th>perc1</th>\n      <th>perc2</th>\n      <th>perc3</th>\n      <th>perc4</th>\n      <th>perc5</th>\n      <th>perc6</th>\n      <th>perc7</th>\n      <th>perc8</th>\n      <th>perc9</th>\n      <th>perc91</th>\n      <th>perc92</th>\n      <th>perc93</th>\n      <th>perc94</th>\n      <th>perc95</th>\n      <th>perc96</th>\n      <th>perc97</th>\n      <th>perc98</th>\n      <th>perc99</th>\n    </tr>\n    <tr>\n      <th>seg_id</th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n      <th></th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>seg_00030f</th>\n      <td>4.491780</td>\n      <td>23.948039</td>\n      <td>0.327904</td>\n      <td>28.836567</td>\n      <td>0.993013</td>\n      <td>0.982353</td>\n      <td>0.970080</td>\n      <td>0.958527</td>\n      <td>0.939587</td>\n      <td>0.910520</td>\n      <td>0.910520</td>\n      <td>0.910520</td>\n      <td>0.864727</td>\n      <td>0.090780</td>\n      <td>0.061067</td>\n      <td>0.061067</td>\n      <td>0.042380</td>\n      <td>0.042380</td>\n      <td>0.030467</td>\n      <td>0.022667</td>\n      <td>0.014080</td>\n      <td>0.006227</td>\n    </tr>\n    <tr>\n      <th>seg_0012b5</th>\n      <td>4.171153</td>\n      <td>35.079793</td>\n      <td>0.295705</td>\n      <td>56.217041</td>\n      <td>0.988540</td>\n      <td>0.976560</td>\n      <td>0.963193</td>\n      <td>0.950827</td>\n      <td>0.930460</td>\n      <td>0.898353</td>\n      <td>0.898353</td>\n      <td>0.898353</td>\n      <td>0.848600</td>\n      <td>0.079640</td>\n      <td>0.055840</td>\n      <td>0.055840</td>\n      <td>0.040540</td>\n      <td>0.040540</td>\n      <td>0.030960</td>\n      <td>0.025073</td>\n      <td>0.017787</td>\n      <td>0.009287</td>\n    </tr>\n    <tr>\n      <th>seg_00184e</th>\n      <td>4.610260</td>\n      <td>48.260349</td>\n      <td>0.428684</td>\n      <td>162.112840</td>\n      <td>0.989680</td>\n      <td>0.979633</td>\n      <td>0.969647</td>\n      <td>0.960153</td>\n      <td>0.944887</td>\n      <td>0.919407</td>\n      <td>0.919407</td>\n      <td>0.919407</td>\n      <td>0.879120</td>\n      <td>0.089073</td>\n      <td>0.060247</td>\n      <td>0.060247</td>\n      <td>0.042600</td>\n      <td>0.042600</td>\n      <td>0.032040</td>\n      <td>0.025053</td>\n      <td>0.017587</td>\n      <td>0.009667</td>\n    </tr>\n    <tr>\n      <th>seg_003339</th>\n      <td>4.531473</td>\n      <td>16.926089</td>\n      <td>0.061889</td>\n      <td>41.240413</td>\n      <td>0.995713</td>\n      <td>0.990387</td>\n      <td>0.984793</td>\n      <td>0.978853</td>\n      <td>0.967093</td>\n      <td>0.944680</td>\n      <td>0.944680</td>\n      <td>0.944680</td>\n      <td>0.905640</td>\n      <td>0.056433</td>\n      <td>0.033640</td>\n      <td>0.033640</td>\n      <td>0.021240</td>\n      <td>0.021240</td>\n      <td>0.014827</td>\n      <td>0.011340</td>\n      <td>0.007700</td>\n      <td>0.003720</td>\n    </tr>\n    <tr>\n      <th>seg_0042cc</th>\n      <td>4.128340</td>\n      <td>33.606882</td>\n      <td>0.073898</td>\n      <td>79.537016</td>\n      <td>0.990227</td>\n      <td>0.978287</td>\n      <td>0.965847</td>\n      <td>0.954340</td>\n      <td>0.934720</td>\n      <td>0.903413</td>\n      <td>0.903413</td>\n      <td>0.903413</td>\n      <td>0.854360</td>\n      <td>0.071340</td>\n      <td>0.049540</td>\n      <td>0.049540</td>\n      <td>0.035853</td>\n      <td>0.035853</td>\n      <td>0.027953</td>\n      <td>0.022607</td>\n      <td>0.015993</td>\n      <td>0.008233</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Pick any model to make predictions and place it where \"rf\" is currently.\n\nrf.fit(scale(dataset[:,:-1]),dataset[:,-1])\n\npredictions = rf.predict(scale(X_test))\n\nsubmission['time_to_failure'] = predictions\nsubmission.to_csv('submission.csv')","execution_count":18,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.6.4","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":1}