'subsample_freq': 0,
'colsample_bytree': 1.0,
'reg_alpha': 0.0,
'reg_lambda': 0.0,
'random_state': 42,
'n_jobs': -1,
'silent': True,
'importance_type': 'split',
'num_boost_round': 500,
'tree_learner': 'feature'
}
lgb = GBM(package='lightgbm',
X=data['X'],
y=data['y'],
feature_names=data['features'],
cv=5,
grid_search=False,
eval_metric='rmse',
parameters=parameters)
lgb.run_model()
print(lgb.__dict__)
'''
print(data['features'])
plot_importance(lgb.model)
plt.show()
plt.clf()
plot_metric(lgb.model, metric='rmse')
plt.show()
plt.clf()
plot_tree(lgb.model)
'cat_features': None,
'grow_policy': None,
'min_data_in_leaf': None,
'min_child_samples': None,
'max_leaves': None,
'num_leaves': None,
'score_function': None,
'leaf_estimation_backtracking': None,
'ctr_history_unit': None,
'monotone_constraints': None
}
catboost = GBM(package='catboost',
X=X[y < 200000],
y=y[y < 200000],
feature_names=data['features'],
cv=5,
grid_search=False,
eval_metric='rmse',
parameters=parameters)
catboost.run_model()
print(catboost.__dict__)
np.save('catboost_res.npy', catboost.__dict__)
catboost.parity_plot(data='train', quantity='CT_RT',
scheme=1).savefig('parity_CT_RT_train.png')
catboost.parity_plot(data='test', quantity='CT_RT',
scheme=1).savefig('parity_CT_RT_test.png')
plt.clf()
explainer = shap.TreeExplainer(catboost.model[-1])
shap_values = explainer.shap_values(data['X'])
'num_boost_round': 5000,
'tree_learner': 'feature'
}
CT_RT = np.array([i[1] for i in data['y2']])
CT_Temp = np.array([i[2] for i in data['y2']])
CT_CS = np.array([i[3] for i in data['y2']])
ID = [i[0] for i in data['y2']]
C = np.array([C_data[i] for i in ID])
lgb = GBM(package='lightgbm',
X=data['X'],
y=data['y'],
model_scheme='LMP',
cv=5,
grid_search=False,
eval_metric='rmse',
parameters=parameters,
CT_RT=CT_RT,
CT_Temp=CT_Temp,
C=C)
lgb.run_model()
print(lgb.__dict__)
lgb.parity_plot(data='train', quantity='LMP').savefig('parity_LMP_train.png')
lgb.parity_plot(data='test', quantity='LMP').savefig('parity_LMP_test.png')
lgb.parity_plot(data='train',
quantity='CT_RT').savefig('parity_CT_RT_train.png')
lgb.parity_plot(data='test', quantity='CT_RT').savefig('parity_CT_RT_test.png')
np.save('lgb_dict.npy', lgb.__dict__)
plt.clf()
'leaf_estimation_backtracking': None,
'ctr_history_unit': None,
'monotone_constraints': None
}
CT_RT = np.array([i[1] for i in data['y2']])
CT_Temp = np.array([i[2] for i in data['y2']])
ID = [i[0] for i in data['y2']]
C = np.array([25 for i in ID])
catboost = GBM(package='catboost',
model_scheme='LMP',
X=data['X'],
y=data['y'],
feature_names=data['features'],
cv=5,
grid_search=False,
eval_metric='rmse',
parameters=parameters,
CT_Temp=CT_Temp,
CT_RT=CT_RT,
C=C)
catboost.run_model()
print(catboost.__dict__)
catboost.parity_plot(data='train',
quantity='LMP').savefig('parity_LMP_train.png')
catboost.parity_plot(data='test',
quantity='LMP').savefig('parity_LMP_test.png')
catboost.parity_plot(data='train',
quantity='CT_RT').savefig('parity_CT_RT_train.png')
catboost.parity_plot(data='test',
'grow_policy': None,
'min_data_in_leaf': None,
'min_child_samples': None,
'max_leaves': None,
'num_leaves': None,
'score_function': None,
'leaf_estimation_backtracking': None,
'ctr_history_unit': None,
'monotone_constraints': None
}
catboost = GBM(package='catboost',
X=data['X'],
y=data['y'],
test_size=0.2,
feature_names=data['features'],
cv=5,
grid_search=False,
eval_metric='rmse',
parameters=parameters)
catboost.run_model()
print(catboost.__dict__)
y_gen = catboost.model[-1].predict(X_generated)
df = pd.DataFrame(data['y'], columns=['rupture life'])
df['sample'] = 'real'
df = df.append(pd.DataFrame({
'rupture life': y_gen,
'sample': ['synthetic' for i in range(len(y_gen))]
}),
'cat_features': None,
'grow_policy': None,
'min_data_in_leaf': None,
'min_child_samples': None,
'max_leaves': None,
'num_leaves': None,
'score_function': None,
'leaf_estimation_backtracking': None,
'ctr_history_unit': None,
'monotone_constraints': None
}
catboost = GBM(package='catboost',
X=data['X'][data['y'] < 200000],
y=data['y'][data['y'] < 200000],
feature_names=data['features'],
cv=5,
grid_search=False,
eval_metric='rmse',
parameters=parameters)
features = [
i for i in df.columns if i not in ['CT_RT', 'CT_CS', 'CT_MCR', 'ID']
]
features = [i for i in features if 'Weighted' not in i]
features.append('CT_RT')
X = df[features].to_numpy(np.float32)
y = df['CT_RT'].to_numpy(np.float32)
pdata = ProcessData(X=X, y=y, features=features)
pdata.clean_data()
data = pdata.get_data()
eval_metric='rmse',
param_grid=parameters_grid)
lgb.run_model()
print(lgb.__dict__)
'''
parameters = {'booster': 'gbtree', #gbtree, gblinear, dart
'eta': 0.3, #learning rate
'gamma': 0, #min split loss
'max_depth': 10,
'tree_method': 'auto'}
xgboost = GBM(package='xgboost',
X=data['X'],
y=data['y'],
feature_names=data['features'],
cv=5,
grid_search=False,
eval_metric='rmse',
parameters=parameters)
xgboost.run_model()
print(xgboost.__dict__)
xgboost.parity_plot(data='train', quantity='LMP').savefig('parity_LMP_train.png')
xgboost.parity_plot(data='test', quantity='LMP').savefig('parity_LMP_test.png')
np.save('xgb_dict.npy', xgboost.__dict__)
plt.clf()
explainer = shap.TreeExplainer(xgboost.model[-1])
shap_values = explainer.shap_values(data['X'])
XX = scale.inverse_transform(data['X'])
'monotone_constraints': None
}
CT_RT = np.array([i[1] for i in data['y2']])
CT_Temp = np.array([i[2] for i in data['y2']])
CT_CS = np.array([i[3] for i in data['y2']])
ID = [i[0] for i in data['y2']]
C = np.array([C_data[i] for i in ID])
catboost = GBM(
package='catboost',
X=data['X'],
y=data['y'],
model_scheme='LMP',
cv=5,
#grid_search=True,
#grid_search_scoring='r2',
#param_grid=parameter_grid,
eval_metric='rmse',
parameters=parameters,
CT_Temp=CT_Temp,
CT_RT=CT_RT,
C=C)
catboost.run_model()
print(catboost.__dict__)
'''
catboost.parity_plot(data='train', quantity='LMP').savefig('parity_LMP_train.png')
catboost.parity_plot(data='test', quantity='LMP').savefig('parity_LMP_test.png')
catboost.parity_plot(data='train', quantity='CT_RT').savefig('parity_CT_RT_train.png')
catboost.parity_plot(data='test', quantity='CT_RT').savefig('parity_CT_RT_test.png')
np.save('catboost_dict.npy', catboost.__dict__)
#'reg_lambda': [None],
#'objective': [None],
#'eta': [None],
#'early_stopping_rounds': [None],
#'cat_features': [None],
'grow_policy': ['SymmetricTree', 'Depthwise', 'Lossguide'],
#'min_data_in_leaf': [None],
#'min_child_samples': [None],
#'max_leaves': [None],
#'num_leaves': [None],
#'score_function': [None],
#'leaf_estimation_backtracking': [None],
#'ctr_history_unit': [None],
#'monotone_constraints': [None]}
}
catboost = GBM(package='catboost',
X=data['X'],
y=data['y'],
feature_names=data['features'],
cv=10,
grid_search=True,
grid_search_scoring='r2',
param_grid=param_grid,
eval_metric='rmse')
catboost.run_model()
print(catboost.__dict__)
np.save('catboost_grid_res.npy', catboost.__dict__)