def predict_sub(model_lgb, testdex, test, subfilename):
print_header('Submission')
print_doing_in_task('predicting')
lgpred = model_lgb.predict(test)
lgsub = pd.DataFrame(lgpred, columns=["deal_probability"], index=testdex)
lgsub['deal_probability'].clip(0.0, 1.0, inplace=True)
print('saving submission file to', subfilename)
lgsub.to_csv(subfilename, index=True, header=True)
print('done')
def prepare_training(mat_filename, dir_feature, predictors, is_textadded):
print_header('Load features')
df, y, len_train, traindex, testdex = load_train_test(['item_id'], TARGET,
DEBUG)
del len_train
gc.collect()
df = drop_col(df, REMOVED_LIST)
# add features
print_doing('add tabular features')
for feature in predictors:
dir_feature_file = dir_feature + feature + '.pickle'
if not os.path.exists(dir_feature_file):
print('can not find {}. Please check'.format(dir_feature_file))
else:
if feature in df:
print('{} already added'.format(feature))
else:
print_doing_in_task('adding {}'.format(feature))
df = add_feature(df, dir_feature_file)
print_memory()
if is_textadded:
# add text_feature
print_doing_in_task('add text features')
ready_df, tfvocab = get_text_matrix(mat_filename, 'all', 2, 0)
# stack
print_doing_in_task('stack')
X = hstack([
csr_matrix(df.loc[traindex, :].values),
ready_df[0:traindex.shape[0]]
]) # Sparse Matrix
testing = hstack([
csr_matrix(df.loc[testdex, :].values), ready_df[traindex.shape[0]:]
])
print_memory()
print_doing_in_task('prepare vocab')
tfvocab = df.columns.tolist() + tfvocab
for shape in [X, testing]:
print("{} Rows and {} Cols".format(*shape.shape))
print("Feature Names Length: ", len(tfvocab))
else:
tfvocab = df.columns.tolist()
testing = hstack([csr_matrix(df.loc[testdex, :].values)])
X = hstack([csr_matrix(df.loc[traindex, :].values)]) # Sparse Matrix
return X, y, testing, tfvocab, df.columns.tolist(), testdex
def get_svdtruncated_vectorizer(todir):
print_doing('doing svdtruncated text feature')
filename = todir + 'text_feature_kernel.pickle'
savename = todir + 'truncated_text_feature_kernel.pickle'
if os.path.exists(savename):
print('done already...')
with open(savename, "rb") as f:
svd_matrix, vocab = pickle.load(f)
with open(filename, "rb") as f:
tfid_matrix, tfvocab = pickle.load(f)
else:
with open(filename, "rb") as f:
tfid_matrix, tfvocab = pickle.load(f)
svdT = TruncatedSVD(n_components=400)
print_doing_in_task('truncated svd')
svd_matrix = svdT.fit_transform(tfid_matrix)
print_doing_in_task('convert to sparse')
svd_matrix = sparse.csr_matrix(svd_matrix, dtype=np.float32)
vocab = []
for i in range(np.shape(svd_matrix)[1]):
vocab.append('lsa' + str(i + 1))
with open(savename, "wb") as f:
pickle.dump((svd_matrix, vocab),
f,
protocol=pickle.HIGHEST_PROTOCOL)
print('---- before truncate')
print(tfid_matrix.shape), print('len of feature:', len(tfvocab))
print('---- after truncate')
print(svd_matrix.shape), print('len of feature:', len(vocab))
if DEBUG:
print(tfid_matrix)
print('\n')
print(svd_matrix)
del svd_matrix, vocab, tfid_matrix, tfvocab
gc.collect()
print_memory()
def train(X, y, num_leave, full_predictors, categorical, predictors,
boosting_type, option, seed):
if DEBUG:
subfilename = '../sub/debug_findseed_{}_{}_{}features_num_leave{}_OPTION{}.csv'. \
format(yearmonthdate_string,boosting_type,str(len(predictors)),num_leave,option)
modelfilename = '../trained_models/debug_findseed_{}_{}_{}features_num_leave{}_OPTION{}.txt'. \
format(yearmonthdate_string,boosting_type,str(len(predictors)),num_leave,option)
else:
subfilename = '../sub/findseed_{}_{}_{}features_num_leave{}_OPTION{}.csv'. \
format(yearmonthdate_string,boosting_type,str(len(predictors)),num_leave,option)
modelfilename = '../trained_models/findseed_{}_{}_{}features_num_leave{}_OPTION{}.txt'. \
format(yearmonthdate_string,boosting_type,str(len(predictors)),num_leave,option)
print_header("Training")
start_time = time.time()
print_doing_in_task('prepare dataset...')
X, X_local_valid, y, y_local_valid = train_test_split(X,
y,
test_size=0.2,
random_state=seed)
X_train, X_valid, y_train, y_valid = train_test_split(X,
y,
test_size=0.10,
random_state=seed)
print('training shape: {} \n'.format(X.shape))
print("Light Gradient Boosting Regressor")
lgbm_params = {
'task': 'train',
'boosting_type': boosting_type,
'objective': 'regression',
'metric': 'rmse',
'max_depth': 15,
'feature_fraction': 0.7,
'bagging_fraction': 0.8,
'learning_rate': 0.1,
'verbose': 0
}
print('params:', lgbm_params)
lgtrain = lgb.Dataset(X_train,
y_train,
feature_name=full_predictors,
categorical_feature=categorical)
lgvalid = lgb.Dataset(X_valid,
y_valid,
feature_name=full_predictors,
categorical_feature=categorical)
if DEBUG:
num_boost_round = 300
early_stopping_rounds = 10
else:
num_boost_round = 20000
early_stopping_rounds = 30
lgb_clf = lgb.train(lgbm_params,
lgtrain,
num_boost_round=num_boost_round,
valid_sets=[lgtrain, lgvalid],
valid_names=['train', 'valid'],
early_stopping_rounds=early_stopping_rounds,
verbose_eval=10)
print_memory()
print_header("Model Report")
runnning_time = '{0:.2f}'.format((time.time() - start_time) / 60)
num_boost_rounds_lgb = lgb_clf.best_iteration
print_doing_in_task('fit val')
val_rmse = '{0:.4f}'.format(
np.sqrt(metrics.mean_squared_error(y_valid, lgb_clf.predict(X_valid))))
print_doing_in_task('fit train')
train_rmse = '{0:.4f}'.format(
np.sqrt(metrics.mean_squared_error(y_train, lgb_clf.predict(X_train))))
print_doing_in_task('fit local val')
local_valid_rmse = '{0:.4f}'.format(
np.sqrt(
metrics.mean_squared_error(y_local_valid,
lgb_clf.predict(X_local_valid))))
diff_lb = '{0:.4f}'.format(abs(float(local_valid_rmse) - 0.2300))
print('OPTION', option)
print('model training time: {} mins'.format(runnning_time))
print('seed number: {}'.format(seed))
print('num_boost_rounds_lgb: {}'.format(num_boost_rounds_lgb))
print('train rmse: {}'.format(train_rmse))
print('val rmse: {}'.format(val_rmse))
print('local valid rmse: {}'.format(local_valid_rmse))
print('diff comapred to lb: {}'.format(diff_lb))
print('saving model to', modelfilename)
lgb_clf.save_model(modelfilename)
seed_name = 'seed_' + str(seed)
LOCAL_VALIDATION_RESULT['seed'][seed_name] = seed
LOCAL_VALIDATION_RESULT['running_time'][seed_name] = runnning_time
LOCAL_VALIDATION_RESULT['num_round'][seed_name] = num_boost_rounds_lgb
LOCAL_VALIDATION_RESULT['train'][seed_name] = train_rmse
LOCAL_VALIDATION_RESULT['val'][seed_name] = val_rmse
LOCAL_VALIDATION_RESULT['local_test'][seed_name] = local_valid_rmse
LOCAL_VALIDATION_RESULT['diff'][seed_name] = diff_lb
return lgb_clf, subfilename
def train(X, y, num_leave, max_depth, full_predictors, categorical, predictors,
boosting_type, option):
print_header("Training")
start_time = time.time()
print_doing_in_task('prepare dataset...')
X_train, X_valid, y_train, y_valid = train_test_split(X,
y,
test_size=0.10,
random_state=SEED)
print('training shape: {} \n'.format(X.shape))
print("Light Gradient Boosting Regressor")
lgbm_params = {
'task': 'train',
'boosting_type': boosting_type,
'objective': 'regression',
'metric': 'rmse',
'max_depth': max_depth,
'num_leave': num_leave,
'feature_fraction': 0.8,
'bagging_fraction': 0.8,
'learning_rate': 0.1,
'lambda_l1': 10,
'max_bin': 512,
'verbose': -1
}
print('params:', lgbm_params)
lgtrain = lgb.Dataset(X_train,
y_train,
feature_name=full_predictors,
categorical_feature=categorical)
lgvalid = lgb.Dataset(X_valid,
y_valid,
feature_name=full_predictors,
categorical_feature=categorical)
if DEBUG:
num_boost_round = 300
early_stopping_rounds = 10
else:
num_boost_round = 20000
early_stopping_rounds = 100
lgb_clf = lgb.train(lgbm_params,
lgtrain,
num_boost_round=num_boost_round,
valid_sets=[lgtrain, lgvalid],
valid_names=['train', 'valid'],
early_stopping_rounds=early_stopping_rounds,
verbose_eval=10)
print_memory()
print_header("Model Report")
runnning_time = '{0:.2f}'.format((time.time() - start_time) / 60)
num_boost_rounds_lgb = lgb_clf.best_iteration
print_doing_in_task('fit val')
val_rmse = '{0:.4f}'.format(
np.sqrt(metrics.mean_squared_error(y_valid, lgb_clf.predict(X_valid))))
print_doing_in_task('fit train')
train_rmse = '{0:.4f}'.format(
np.sqrt(metrics.mean_squared_error(y_train, lgb_clf.predict(X_train))))
print_header("Model Report")
print('boosting_type {}, num_leave {}, max_depth {}'.format(
boosting_type, num_leave, max_depth))
print('model training time: {0:.2f} mins'.format(
(time.time() - start_time) / 60))
print('num_boost_rounds_lgb: {}'.format(lgb_clf.best_iteration))
print('best rmse: {0:.4f}'.format(
np.sqrt(metrics.mean_squared_error(y_valid,
lgb_clf.predict(X_valid)))))
model = '{}_{}_{}'.format(boosting_type, num_leave, max_depth)
LOCAL_TUNE_RESULT['running_time'][model] = runnning_time
LOCAL_TUNE_RESULT['num_round'][model] = num_boost_rounds_lgb
LOCAL_TUNE_RESULT['train'][model] = train_rmse
LOCAL_TUNE_RESULT['val'][model] = val_rmse