def score_solution():
'''
this function fits the pipeline to training data and tests the model on the testing data and finally returns roc_auc_score
'''
import solution
pipeline = solution.get_pipeline()
error_message = 'Your `solution.get_pipeline` implementation should ' 'return an `sklearn.pipeline.Pipeline`.'
assert isinstance(pipeline, sklearn.pipeline.Pipeline), error_message
# Train the model on the training DataFrame.
X_train, y_train = get_data(subset='train')
X_train, _ = encoding(missing_values(X_train))
print('\n')
print('Training........')
print('\n')
pipeline.fit(X_train, y_train)
# Apply the model to the test DataFrame.
X_test, y_test = get_data(subset='test')
X_test, _ = encoding(missing_values(X_test))
print('\n')
print('Test Results')
print('\n')
y_pred = pipeline.predict_proba(X_test)
assert (y_pred.ndim == 1) or (
y_pred.ndim == 2 and y_pred.shape[1] == 2
), 'The predicted probabilities should match sklearn' 's ' '`predict_proba` output shape.'
y_pred = y_pred if y_pred.ndim == 1 else y_pred[:, 1]
return sklearn.metrics.roc_auc_score(y_test, y_pred)
def score_solution(model, save=0):
'''
Added a model and save parameter:
model ~ hold a classification model
save ~ Flag used to save the best model on file using jobLib
'''
# Ask the solution for the model pipeline.
import solution
pipeline = solution.get_pipeline(model)
error_message = 'Your `solution.get_pipeline` implementation should ' \
'return an `sklearn.pipeline.Pipeline`.'
assert isinstance(pipeline, sklearn.pipeline.Pipeline), error_message
# Train the model on the training DataFrame.
X_train, y_train = get_data(subset='train')
pipeline.fit(X_train, y_train)
# Apply the model to the test DataFrame.
X_test, y_test = get_data(subset='test')
y_pred = pipeline.predict_proba(X_test)
# Check that the predicted probabilities have an sklearn-compatible shape.
assert (y_pred.ndim == 1) or \
(y_pred.ndim == 2 and y_pred.shape[1] == 2), \
'The predicted probabilities should match sklearn''s ' \
'`predict_proba` output shape`.'
y_pred = y_pred if y_pred.ndim == 1 else y_pred[:, 1]
# Evaluate the predictions with the AUC of the ROC curve.
if (save == 1): joblib.dump(pipeline, 'Best_Estimator.sav')
return sklearn.metrics.roc_auc_score(y_test, y_pred)
def grid_search():
from solution import get_pipeline
s = settings()
grid_params = s['GRID_PARAMS']
pipeline = get_pipeline()
search = GridSearchCV(pipeline, grid_params, n_jobs=8)
X_train, y_train = get_data(subset='train')
results = search.fit(X_train, y_train)
print(search.best_params_)
print("====='")
print(search.cv_results_)
def score_solution():
# Ask the solution for the model pipeline.
import solution
pipeline = solution.get_pipeline()
error_message = 'Your `solution.get_pipeline` implementation should ' \
'return an `sklearn.pipeline.Pipeline`.'
assert isinstance(pipeline, sklearn.pipeline.Pipeline), error_message
# Train the model on the training DataFrame.
X_train, y_train = get_data(subset='train')
pipeline.fit(X_train, y_train)
# Apply the model to the test DataFrame.
X_test, y_test = get_data(subset='test')
y_pred = pipeline.predict_proba(X_test)
# Check that the predicted probabilities have an sklearn-compatible shape.
assert (y_pred.ndim == 1) or \
(y_pred.ndim == 2 and y_pred.shape[1] == 2), \
'The predicted probabilities should match sklearn''s ' \
'`predict_proba` output shape.'
y_pred = y_pred if y_pred.ndim == 1 else y_pred[:, 1]
# Evaluate the predictions with the AUC of the ROC curve.
return sklearn.metrics.roc_auc_score(y_test, y_pred)
# param_grid = dict(
# #pipeline__missingcategoricalstransformer__strategy=["none"],
# tensorflowestimator__dropout=[0.15, 0.20, 0.25, 0.3, 0.35, 0.40],
# tensorflowestimator__hidden_units=[[64,32], [24], [48], [30, 12], [128, 48], [128,48,12]],
# tensorflowestimator__training_steps=[900]
# )
# TensorFlowEstimator(dropout=0.3, hidden_units=[128, 48, 12], training_steps=900)
param_grid = dict(
#pipeline__missingcategoricalstransformer__strategy=["none"],
tensorflowestimator__dropout=[0.35, 0.40, 0.45, 0.5],
tensorflowestimator__hidden_units=[[128,48,12], [1024, 514, 256, 128, 64], [200, 100, 40], [56, 28, 12]],
tensorflowestimator__training_steps=[900]
)
#{'tensorflowestimator__dropout': 0.4, 'tensorflowestimator__hidden_units': [200, 100, 40], 'tensorflowestimator__training_steps': 900}
gs = GridSearchCV(solution.get_pipeline(), param_grid, cv=6, n_jobs=4)
from challenge import get_dataw
X, y = get_data()
# print(X.index.values)
# index = np.arange(X.index.values[0], X.index.values[-1]+1)
# print(index)
#
# X.set_index(index, inplace=True)
# print(X.index.values)
gs.fit(X, y)
print(gs.best_params_)