def cross_validation_models(models_set, X_train_set, y_train_set):
scores = []
names = []
for name, model in models_set:
names.append(name)
scores.append(
np.sqrt(pp.score_model(model, X_train_set, y_train_set)).mean())
tab = pd.DataFrame({"Model": names, "Score": scores})
tab = tab.sort_values(by=['Score'], ascending=True)
return (tab)
def get_validation_scores(models, X_train, y_train, X_test = [], y_test = []):
scores_val = []
scores_test = []
names = []
for name, model in models:
names.append(name)
scores_val.append(np.sqrt(pp.score_model(model, X_train, y_train)).mean())
if len(X_test) != 0:
model.fit(X_train, y_train)
st = get_test_scores(model, X_test, y_test)
scores_test.append(st)
if len(X_test) != 0:
tab = pd.DataFrame({ "Model" : names, "Cross Validation" : scores_val, "Test": scores_test })
tab = tab.sort_values(by=['Test'], ascending = True)
else:
tab = pd.DataFrame({ "Model" : names, "Cross Validation" : scores_val })
tab = tab.sort_values(by=['Cross Validation'], ascending = True)
return(tab)
def get_validation_scores(models,
X_train,
y_train,
folds,
X_test=[],
y_test=[]):
scores_val_mean = []
scores_val_std = []
scores_val = []
scores_test = []
names = []
for name, model in models:
names.append(name)
val_scores = np.sqrt(
pp.score_model(model, X_train, y_train, n_folds=folds))
scores_val.append(val_scores)
scores_val_mean.append(val_scores.mean())
scores_val_std.append(val_scores.std())
if len(X_test) != 0:
model.fit(X_train, y_train)
st = get_test_scores(model, X_test, y_test)
scores_test.append(st)
if len(X_test) != 0:
tab = pd.DataFrame({
"Model": names,
"Cross Validation (Mean)": scores_val_mean,
"Cross Validation (Std)": scores_val_std,
"Cross Validation (Scores)": scores_val,
"Test": scores_test
})
tab.sort_values(by=['Test'], ascending=True, inplace=True)
else:
tab = pd.DataFrame({
"Model": names,
"Cross Validation (Mean)": scores_val_mean,
"Cross Validation (Std)": scores_val_std,
"Cross Validation (Scores)": scores_val
})
tab.sort_values(by=['Cross Validation (Mean)'],
ascending=True,
inplace=True)
return (tab)
models.append(("byr", model_byr))
models.append(("rforest", model_rforest))
models.append(("xgb", model_xgb))
models.append(("GBoost", model_GBoost))
models.append(("lgb", model_lgb))
models.append(("lasso_lars", model_lasso_lars))
models.append(("lsvr", model_lsvr))
models.append(("sgd", model_sgd))
models.append(("extra", model_extra))
scores = []
names = []
for name, model in models:
names.append(name)
scores.append(
np.sqrt(pp.score_model(model, train_X_reduced, train_y)).mean())
tab = pd.DataFrame({"Model": names, "Score": scores})
tab = tab.sort_values(by=['Score'], ascending=True)
print(tab)
for model in models:
model.fit(train_X, train_y)
averaged_models = em.AveragingModels(
models=[model_svr, model_ENet, model_KRR, model_xgb])
score_avg = np.sqrt(pp.score_model(averaged_models, train_X_reduced, train_y))
print(" Averaged base models score: {:.6f} ({:.6f})\n".format(
score_avg.mean(), score_avg.std()))
averaged_models.fit(train_X_reduced, train_y)
file_name='submission.csv'):
model.fit(X_train_set, y_train_set)
predicted = np.expm1(model.predict(test_X_to_submit))
print(predicted)
submission = pd.DataFrame({'Id': ids, 'SalePrice': predicted})
submission.to_csv(file_name, index=False)
train_submit(em.AveragingModels(models=[model_svr, model_byr, model_xgb]),
train_X_reduced, train_y, test_X_reduced, 'submission_avg_1.csv')
# Best one so far: model_svr, model_KRR, model_xgb
# Try this combination: model_svr, model_KRR, averaged_models_tree
averaged_models = em.AveragingModels(models=[model_svr, model_KRR, model_xgb])
score_avg = np.sqrt(pp.score_model(averaged_models, train_X_reduced, train_y))
print(" Averaged base models score: {:.6f} ({:.6f})\n".format(
score_avg.mean(), score_avg.std()))
stacked_averaged_models = em.StackingAveragedModels(
base_models=[model_KRR, model_xgb], meta_model=model_svr)
score_stacked_averaged = np.sqrt(
pp.score_model(stacked_averaged_models, train_X_reduced, train_y))
print(" Stacked Averaged base models score: {:.6f} ({:.6f})\n".format(
score_stacked_averaged.mean(), score_stacked_averaged.std()))
stacked_averaged_models.fit(train_X_reduced, train_y)
predicted_prices_stacked_averaged = np.expm1(
stacked_averaged_models.predict(test_X_reduced))
print(predicted_prices_stacked_averaged)