In [1]:
import os
import pandas as pd
import numpy as np
import gc
import lightgbm as lgb
from sklearn.metrics import mean_squared_error
try:
import cPickle as pickle
except:
import pickle
import time
from scipy.stats import skew, kurtosis
In [2]:
DATA_PATH = 'data/'
PRED_TEST_PATH = ''
PRED_TRAIN_PATH = ''
FOLDS_PATH = ''
MODEL_NAME = 'izmajlovkonstantin'
In [3]:
RANDOM_STATE = 22
np.random.seed(RANDOM_STATE)
In [4]:
train_df = pd.read_csv(f'{DATA_PATH}train.csv')
test_df = pd.read_csv(f'{DATA_PATH}test.csv')
In [5]:
y = np.log1p(train_df.target.values)
y.shape
Out[5]:
(4459,)
In [6]:
columns_to_use = test_df.columns.tolist()
del columns_to_use[0] # Remove 'ID'
X = train_df[columns_to_use]
test = test_df[columns_to_use]
In [7]:
constant_columns = train_df.loc[:, (train_df == train_df.iloc[0]).all()].columns.tolist()
correlated_columns = ['bba402827',
'd60ddde1b',
'33ed23348',
'912836770',
'acc5b709d',
'f8d75792f',
'22c933b9b',
'f333a5f60']
print(len(constant_columns))
256
In [8]:
def find_too_freq_values(thresh=None, constant_value=0):
cols_with_too_freq_values= []
if thresh is None:
thresh =0.98
for column in train_df.columns:
counts = train_df[column].value_counts()
try:
counts[constant_value]
except KeyError:
continue
value_fraction = counts[constant_value] / len(train_df)
if value_fraction > thresh:
cols_with_too_freq_values.append(column)
return cols_with_too_freq_values
In [9]:
%%time
cols_with_too_freq_values = find_too_freq_values()
print(len(cols_with_too_freq_values))
2870
Wall time: 3.96 s
In [10]:
cols_to_remove = list(set(constant_columns)|set(correlated_columns))
len(cols_to_remove)
Out[10]:
264
In [11]:
columns_to_use = test_df.columns.tolist()[1:] # Remove 'ID'
columns_to_use = [x for x in columns_to_use if x not in cols_to_remove]
In [12]:
X = train_df[columns_to_use]
test = test_df[columns_to_use]
In [13]:
from sklearn.decomposition import PCA, TruncatedSVD, FastICA, FactorAnalysis
from sklearn.random_projection import SparseRandomProjection, GaussianRandomProjection
In [14]:
random_state = 13
feat_extractors_dict = {'pca': PCA(n_components=0.9, random_state=random_state),
'tsvd': TruncatedSVD(n_components = 50, n_iter=10, random_state=random_state),
'fa': FactorAnalysis(n_components=50, random_state=random_state),
'gauss': GaussianRandomProjection(n_components=50, eps=0.1, random_state=random_state),
'srp': SparseRandomProjection(n_components=50, random_state=random_state)
}
def create_dim_reduction_feats(df, train=False):
full_X_arr = []
all_cols = []
if train:
for k, v in feat_extractors_dict.items():
print(f'Process {k}')
X_arr = v.fit_transform(df)
n_components = X_arr.shape[1]
curr_cols = [str(k)+f'{i}' for i in range(n_components)]
all_cols += curr_cols
if len(full_X_arr) ==0:
full_X_arr = X_arr
else:
full_X_arr = np.hstack((full_X_arr, X_arr))
else:
for k, v in feat_extractors_dict.items():
print(f'Process {k}')
X_arr = v.transform(df)
n_components = X_arr.shape[1]
curr_cols = [str(k)+f'{i}' for i in range(n_components)]
all_cols += curr_cols
if len(full_X_arr) ==0:
full_X_arr = X_arr
else:
full_X_arr = np.hstack((full_X_arr, X_arr))
new_df = pd.DataFrame(full_X_arr, columns=all_cols, index=df.index)
return new_df
In [15]:
%%time
train_dim_reduction = create_dim_reduction_feats(X, train=True)
test_dim_reduction = create_dim_reduction_feats(test)
Process pca
Process tsvd
Process fa
Process gauss
Process srp
Process pca
Process tsvd
Process fa
Process gauss
Process srp
Wall time: 1min 59s
In [20]:
def aggregate_row(row):
non_zero_values = row.iloc[row.nonzero()]
# print(non_zero_values)
if len(non_zero_values.value_counts())>1:
aggs = {'non_zero_mean': non_zero_values.mean(),
'non_zero_std': non_zero_values.std(),
'non_zero_max': non_zero_values.max(),
'non_zero_min': non_zero_values.min(),
'non_zero_sum': non_zero_values.sum(),
'non_zero_skewness': skew(non_zero_values),
'non_zero_kurtosis': kurtosis(non_zero_values),
'non_zero_median': non_zero_values.median(),
'non_zero_q1': np.percentile(non_zero_values, q=25),
'non_zero_q3': np.percentile(non_zero_values, q=75),
'non_zero_log_mean': np.log1p(non_zero_values.astype('float64')).mean(),
'non_zero_log_std': np.log1p(non_zero_values.astype('float64')).std(),
'non_zero_log_max': np.log1p(non_zero_values.astype('float64')).max(),
'non_zero_log_min': np.log1p(non_zero_values.astype('float64')).min(),
'non_zero_log_sum': np.log1p(non_zero_values.astype('float64')).sum(),
'non_zero_log_skewness': skew(np.log1p(non_zero_values.astype('float64'))),
'non_zero_log_kurtosis': kurtosis(np.log1p(non_zero_values.astype('float64'))) ,
'non_zero_log_median': np.log1p(non_zero_values.astype('float64')).median(),
'non_zero_log_q1': np.percentile(np.log1p(non_zero_values.astype('float64')), q=25),
'non_zero_log_q3': np.percentile(np.log1p(non_zero_values.astype('float64')), q=75),
'non_zero_count': non_zero_values.count(),
'non_zero_fraction': non_zero_values.count() / row.count()
}
else:
aggs = {'non_zero_mean': -999999,
'non_zero_std':-999999,
'non_zero_max': -999999,
'non_zero_min': -999999,
'non_zero_sum': -999999,
'non_zero_skewness': -999999,
'non_zero_kurtosis': -999999,
'non_zero_median': -999999,
'non_zero_q1': -999999,
'non_zero_q3':-999999,
'non_zero_log_mean': -999999,
'non_zero_log_std': -999999,
'non_zero_log_max': -999999,
'non_zero_log_min': -999999,
'non_zero_log_sum': -999999,
'non_zero_log_skewness': -999999,
'non_zero_log_kurtosis': -999999,
'non_zero_log_median': -999999,
'non_zero_log_q1': -999999,
'non_zero_log_q3':-999999,
'non_zero_count': -999999,
'non_zero_fraction': -999999
}
return pd.Series(aggs)
In [21]:
def transform(X):
X_agg = X.apply(aggregate_row, axis=1)
return X_agg
In [22]:
%%time
df_with_row_statistic_train = transform(train_df.iloc[:, 2:])
df_with_row_statistic_test = transform(test_df.iloc[:, 1:])
Wall time: 3min 33s
In [23]:
X_n = pd.concat((X,df_with_row_statistic_train),axis = 1)
X_n = pd.concat((X_n,train_dim_reduction),axis = 1)
In [24]:
test_n = pd.concat((test,df_with_row_statistic_test),axis = 1)
test_n = pd.concat((test_n,test_dim_reduction),axis = 1)
In [25]:
def generate_features(dframe,df):
interactions = []
for i in dframe.columns:
for j in dframe.columns:
if i==j:
continue
else:
list_of_indexes = [i,j]
list_of_indexes.sort()
if list_of_indexes not in interactions:
interactions.append(list_of_indexes)
for cols in interactions:
col1 = cols[0]
col2 = cols[1]
name = col1 + "/" + (col2)
feature_interactions.append(name)
df = pd.concat([df, pd.Series(dframe.loc[:,col1] /(1+ dframe.loc[:,col2]), name=name)], axis=1)
return df
log_feats =['26ab20ff9',
'6786ea46d',
'15ace8c9f',
'b6fa5a5fd',
'a72e0bf30',
'fb387ea33',
'f190486d6',
'9fd594eec',
'251d1aa17',
'1c71183bb',
'5e1085022',
'5c6487af1',
'b58127585',
'37f57824c',
'3bdee45be',
'08e89cc54',
'3e1100230',
'91f701ba2',
'66ace2992',
'b791ce9aa']
logic = X_n.loc[:, log_feats]
feature_interactions = []
X_n = generate_features(logic,X_n)
logic = test_n.loc[:, log_feats]
feature_interactions = []
test_n = generate_features(logic,test_n)
In [54]:
from boostaroota import BoostARoota
br = BoostARoota(metric='rmse')
br.fit(X_n, y)
Round: 1 iteration: 1
Round: 1 iteration: 2
Round: 1 iteration: 3
Round: 1 iteration: 4
Round: 1 iteration: 5
Round: 1 iteration: 6
Round: 1 iteration: 7
Round: 1 iteration: 8
Round: 1 iteration: 9
Round: 1 iteration: 10
Round: 2 iteration: 1
Round: 2 iteration: 2
Round: 2 iteration: 3
Round: 2 iteration: 4
Round: 2 iteration: 5
Round: 2 iteration: 6
Round: 2 iteration: 7
Round: 2 iteration: 8
Round: 2 iteration: 9
Round: 2 iteration: 10
Round: 3 iteration: 1
Round: 3 iteration: 2
Round: 3 iteration: 3
Round: 3 iteration: 4
Round: 3 iteration: 5
Round: 3 iteration: 6
Round: 3 iteration: 7
Round: 3 iteration: 8
Round: 3 iteration: 9
Round: 3 iteration: 10
Round: 4 iteration: 1
Round: 4 iteration: 2
Round: 4 iteration: 3
Round: 4 iteration: 4
Round: 4 iteration: 5
Round: 4 iteration: 6
Round: 4 iteration: 7
Round: 4 iteration: 8
Round: 4 iteration: 9
Round: 4 iteration: 10
BoostARoota ran successfully! Algorithm went through 4 rounds.
Out[54]:
<boostaroota.boostaroota.BoostARoota at 0x1e2291d1748>
In [55]:
remaining_vars = list(br.keep_vars_)
In [75]:
pd.DataFrame(remaining_vars).to_csv('remaining_vars_4.csv',index = False)
In [49]:
def run_lgb(train_X, train_y, val_X, val_y, test_X):
params = {
"objective" : "regression",
"metric" : "rmse",
"num_leaves" : 40,
'max_depth': 8, # -1,
"learning_rate" : 0.005,
"bagging_fraction" : 0.7,
"feature_fraction" : 0.1, # 0.6,
"bagging_frequency" : 6,
"bagging_seed" : 44,
"verbosity" : -1,
'num_threads' : 4,
"seed": 44,
"nthread" : 5
}
start_time = time.time()
lgtrain = lgb.Dataset(train_X, label=train_y)
lgval = lgb.Dataset(val_X, label=val_y)
model = lgb.train(params, lgtrain, 5000,
valid_sets=[lgtrain, lgval],
early_stopping_rounds=100,
verbose_eval=150)
print('Model training done in {} seconds.'.format(time.time() - start_time))
pred_test_y = np.expm1(model.predict(test_X, num_iteration=model.best_iteration))
pred_oof_log = model.predict(val_X, num_iteration=model.best_iteration)
return pred_test_y, pred_oof_log, model
In [50]:
def run_calculations(X, test, big_cv_folds, func_name = None):
if not func_name:
return print('The function to run is not defined')
else:
y_oof_20_preds = []
fold_errors_20_preds =[]
avg_test_pred_20_preds = []
for ind, cv_folds in enumerate(big_cv_folds):
print('Fitting big fold', ind+1, 'out of', len(big_cv_folds))
y_oof = np.zeros((y.shape[0]))
fold_errors =[]
pred_test_list = []
for i, (train_index, val_index) in enumerate(cv_folds):
print('Fitting sub fold', i+1, 'out of', len(cv_folds))
X_train, X_val = X.iloc[train_index], X.iloc[val_index]
y_train, y_val = y[train_index], y[val_index]
# part to include additional functions
if func_name == 'lgb':
pred_test_y, pred_oof_log, clf = run_lgb(X_train, y_train, X_val, y_val, test)
else:
return print('The function to run is not correct')
y_oof[val_index] = pred_oof_log
curr_fe = np.sqrt(mean_squared_error(y_val, pred_oof_log))
print(f'Fold error {curr_fe}')
fold_errors.append(curr_fe)
pred_test_list.append(list(pred_test_y))
print('Total error', np.sqrt(mean_squared_error(y, y_oof)))
total_fe_std = round(np.std(fold_errors), 5)
print(f'Total std {total_fe_std}')
avg_test_pred = np.mean(pred_test_list, axis=0)
avg_test_pred_20_preds.append(avg_test_pred)
fold_errors_20_preds.append(fold_errors)
y_oof_20_preds.append(y_oof)
return y_oof_20_preds, avg_test_pred_20_preds, fold_errors_20_preds
In [45]:
with open(f'{FOLDS_PATH}custom_cv.pkl', 'rb') as f:
cv_folds = pickle.load(f)
In [63]:
X_n[remaining_vars].shape
Out[63]:
(4459, 460)
In [58]:
%%time
y_oof_lgb, pred_test_list_lgb, fold_errors = run_calculations(X_n[remaining_vars], test_n[remaining_vars], cv_folds, 'lgb')
Fitting big fold 1 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.37333 valid_1's rmse: 1.48188
[300] training's rmse: 1.16663 valid_1's rmse: 1.3859
[450] training's rmse: 1.03466 valid_1's rmse: 1.35263
[600] training's rmse: 0.93747 valid_1's rmse: 1.34076
[750] training's rmse: 0.855625 valid_1's rmse: 1.33541
[900] training's rmse: 0.785514 valid_1's rmse: 1.33193
[1050] training's rmse: 0.726519 valid_1's rmse: 1.33053
[1200] training's rmse: 0.670614 valid_1's rmse: 1.32983
[1350] training's rmse: 0.623805 valid_1's rmse: 1.32913
[1500] training's rmse: 0.580083 valid_1's rmse: 1.32845
Early stopping, best iteration is:
[1497] training's rmse: 0.580998 valid_1's rmse: 1.32842
Model training done in 13.497329235076904 seconds.
Fold error 1.3284191745053764
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40385 valid_1's rmse: 1.4507
[300] training's rmse: 1.22758 valid_1's rmse: 1.33725
[450] training's rmse: 1.11926 valid_1's rmse: 1.29462
[600] training's rmse: 1.04105 valid_1's rmse: 1.27785
[750] training's rmse: 0.97653 valid_1's rmse: 1.26914
[900] training's rmse: 0.91944 valid_1's rmse: 1.26439
[1050] training's rmse: 0.868298 valid_1's rmse: 1.26178
[1200] training's rmse: 0.820779 valid_1's rmse: 1.26015
[1350] training's rmse: 0.777281 valid_1's rmse: 1.25884
[1500] training's rmse: 0.737844 valid_1's rmse: 1.25755
[1650] training's rmse: 0.69949 valid_1's rmse: 1.25694
Early stopping, best iteration is:
[1580] training's rmse: 0.717114 valid_1's rmse: 1.25664
Model training done in 17.597947597503662 seconds.
Fold error 1.2566374444854034
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40023 valid_1's rmse: 1.46273
[300] training's rmse: 1.21981 valid_1's rmse: 1.36349
[450] training's rmse: 1.11023 valid_1's rmse: 1.33023
[600] training's rmse: 1.03148 valid_1's rmse: 1.31928
[750] training's rmse: 0.965883 valid_1's rmse: 1.31337
[900] training's rmse: 0.908509 valid_1's rmse: 1.31157
[1050] training's rmse: 0.855353 valid_1's rmse: 1.31013
[1200] training's rmse: 0.806325 valid_1's rmse: 1.30818
[1350] training's rmse: 0.761642 valid_1's rmse: 1.30831
Early stopping, best iteration is:
[1299] training's rmse: 0.77714 valid_1's rmse: 1.3075
Model training done in 14.258858680725098 seconds.
Fold error 1.3074992591375598
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.38924 valid_1's rmse: 1.52266
[300] training's rmse: 1.20784 valid_1's rmse: 1.42601
[450] training's rmse: 1.09623 valid_1's rmse: 1.39001
[600] training's rmse: 1.0153 valid_1's rmse: 1.375
[750] training's rmse: 0.947752 valid_1's rmse: 1.3671
[900] training's rmse: 0.889397 valid_1's rmse: 1.36209
[1050] training's rmse: 0.83632 valid_1's rmse: 1.35929
[1200] training's rmse: 0.78864 valid_1's rmse: 1.35729
[1350] training's rmse: 0.74396 valid_1's rmse: 1.35586
[1500] training's rmse: 0.703655 valid_1's rmse: 1.35491
[1650] training's rmse: 0.665575 valid_1's rmse: 1.35425
[1800] training's rmse: 0.630312 valid_1's rmse: 1.35326
[1950] training's rmse: 0.596747 valid_1's rmse: 1.35251
Early stopping, best iteration is:
[1890] training's rmse: 0.609615 valid_1's rmse: 1.35233
Model training done in 19.519783973693848 seconds.
Fold error 1.3523329032290727
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39373 valid_1's rmse: 1.48398
[300] training's rmse: 1.208 valid_1's rmse: 1.39129
[450] training's rmse: 1.09428 valid_1's rmse: 1.35677
[600] training's rmse: 1.01125 valid_1's rmse: 1.34367
[750] training's rmse: 0.942241 valid_1's rmse: 1.33742
[900] training's rmse: 0.880986 valid_1's rmse: 1.33389
[1050] training's rmse: 0.82627 valid_1's rmse: 1.33084
[1200] training's rmse: 0.775344 valid_1's rmse: 1.32933
[1350] training's rmse: 0.728423 valid_1's rmse: 1.32781
[1500] training's rmse: 0.686399 valid_1's rmse: 1.3274
[1650] training's rmse: 0.6467 valid_1's rmse: 1.32638
[1800] training's rmse: 0.610809 valid_1's rmse: 1.32589
[1950] training's rmse: 0.575863 valid_1's rmse: 1.32569
[2100] training's rmse: 0.54447 valid_1's rmse: 1.32468
[2250] training's rmse: 0.515495 valid_1's rmse: 1.32446
Early stopping, best iteration is:
[2184] training's rmse: 0.528211 valid_1's rmse: 1.32434
Model training done in 18.735880851745605 seconds.
Fold error 1.3243421378894245
Total error 1.3216567175702476
Total std 0.03199
Fitting big fold 2 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.36052 valid_1's rmse: 1.49942
[300] training's rmse: 1.15065 valid_1's rmse: 1.4128
[450] training's rmse: 1.01893 valid_1's rmse: 1.38455
[600] training's rmse: 0.923591 valid_1's rmse: 1.37502
[750] training's rmse: 0.846289 valid_1's rmse: 1.37107
[900] training's rmse: 0.778318 valid_1's rmse: 1.36889
[1050] training's rmse: 0.717233 valid_1's rmse: 1.368
[1200] training's rmse: 0.662616 valid_1's rmse: 1.36721
[1350] training's rmse: 0.612582 valid_1's rmse: 1.36677
Early stopping, best iteration is:
[1317] training's rmse: 0.62305 valid_1's rmse: 1.3666
Model training done in 10.660483121871948 seconds.
Fold error 1.3665990652385995
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40513 valid_1's rmse: 1.4435
[300] training's rmse: 1.22861 valid_1's rmse: 1.32917
[450] training's rmse: 1.11985 valid_1's rmse: 1.28642
[600] training's rmse: 1.04128 valid_1's rmse: 1.2706
[750] training's rmse: 0.975939 valid_1's rmse: 1.26296
[900] training's rmse: 0.918592 valid_1's rmse: 1.25882
[1050] training's rmse: 0.866015 valid_1's rmse: 1.25616
[1200] training's rmse: 0.818539 valid_1's rmse: 1.25518
Early stopping, best iteration is:
[1204] training's rmse: 0.81727 valid_1's rmse: 1.25501
Model training done in 13.68040418624878 seconds.
Fold error 1.2550137531480263
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40274 valid_1's rmse: 1.46324
[300] training's rmse: 1.22536 valid_1's rmse: 1.34945
[450] training's rmse: 1.11682 valid_1's rmse: 1.30577
[600] training's rmse: 1.03713 valid_1's rmse: 1.28875
[750] training's rmse: 0.97146 valid_1's rmse: 1.27762
[900] training's rmse: 0.913814 valid_1's rmse: 1.27142
[1050] training's rmse: 0.861136 valid_1's rmse: 1.26831
[1200] training's rmse: 0.812858 valid_1's rmse: 1.2652
[1350] training's rmse: 0.769093 valid_1's rmse: 1.26393
Early stopping, best iteration is:
[1269] training's rmse: 0.792155 valid_1's rmse: 1.26336
Model training done in 14.86922574043274 seconds.
Fold error 1.2633560256328378
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39088 valid_1's rmse: 1.51715
[300] training's rmse: 1.2094 valid_1's rmse: 1.41606
[450] training's rmse: 1.09864 valid_1's rmse: 1.3804
[600] training's rmse: 1.01802 valid_1's rmse: 1.36725
[750] training's rmse: 0.950956 valid_1's rmse: 1.36128
[900] training's rmse: 0.891146 valid_1's rmse: 1.35947
[1050] training's rmse: 0.838276 valid_1's rmse: 1.35748
[1200] training's rmse: 0.789722 valid_1's rmse: 1.35598
[1350] training's rmse: 0.744647 valid_1's rmse: 1.35514
[1500] training's rmse: 0.703253 valid_1's rmse: 1.35526
Early stopping, best iteration is:
[1405] training's rmse: 0.728969 valid_1's rmse: 1.35483
Model training done in 14.59395146369934 seconds.
Fold error 1.3548337762027558
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39715 valid_1's rmse: 1.45332
[300] training's rmse: 1.21175 valid_1's rmse: 1.35437
[450] training's rmse: 1.09784 valid_1's rmse: 1.3219
[600] training's rmse: 1.01413 valid_1's rmse: 1.31127
[750] training's rmse: 0.945557 valid_1's rmse: 1.30642
[900] training's rmse: 0.885117 valid_1's rmse: 1.30376
[1050] training's rmse: 0.830751 valid_1's rmse: 1.30241
[1200] training's rmse: 0.780908 valid_1's rmse: 1.30117
Early stopping, best iteration is:
[1176] training's rmse: 0.788456 valid_1's rmse: 1.30085
Model training done in 12.663126230239868 seconds.
Fold error 1.3008496250324584
Total error 1.327615896386753
Total std 0.04578
Fitting big fold 3 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.37842 valid_1's rmse: 1.47649
[300] training's rmse: 1.17211 valid_1's rmse: 1.3815
[450] training's rmse: 1.04226 valid_1's rmse: 1.3484
[600] training's rmse: 0.945393 valid_1's rmse: 1.33613
[750] training's rmse: 0.865403 valid_1's rmse: 1.33119
[900] training's rmse: 0.795207 valid_1's rmse: 1.32875
[1050] training's rmse: 0.732307 valid_1's rmse: 1.32736
[1200] training's rmse: 0.675215 valid_1's rmse: 1.32683
Early stopping, best iteration is:
[1161] training's rmse: 0.689209 valid_1's rmse: 1.32638
Model training done in 11.151170253753662 seconds.
Fold error 1.3263829505300186
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40398 valid_1's rmse: 1.45642
[300] training's rmse: 1.22685 valid_1's rmse: 1.34911
[450] training's rmse: 1.11922 valid_1's rmse: 1.30974
[600] training's rmse: 1.0412 valid_1's rmse: 1.29533
[750] training's rmse: 0.976571 valid_1's rmse: 1.28704
[900] training's rmse: 0.919017 valid_1's rmse: 1.2846
[1050] training's rmse: 0.868098 valid_1's rmse: 1.28208
[1200] training's rmse: 0.821151 valid_1's rmse: 1.28082
[1350] training's rmse: 0.778874 valid_1's rmse: 1.28041
[1500] training's rmse: 0.738559 valid_1's rmse: 1.27957
[1650] training's rmse: 0.700158 valid_1's rmse: 1.27834
Early stopping, best iteration is:
[1656] training's rmse: 0.698649 valid_1's rmse: 1.27805
Model training done in 16.427213191986084 seconds.
Fold error 1.278050144096793
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39531 valid_1's rmse: 1.50881
[300] training's rmse: 1.21663 valid_1's rmse: 1.4096
[450] training's rmse: 1.10788 valid_1's rmse: 1.37343
[600] training's rmse: 1.02838 valid_1's rmse: 1.35817
[750] training's rmse: 0.962527 valid_1's rmse: 1.35164
[900] training's rmse: 0.905084 valid_1's rmse: 1.34713
[1050] training's rmse: 0.853239 valid_1's rmse: 1.34494
[1200] training's rmse: 0.804866 valid_1's rmse: 1.34358
[1350] training's rmse: 0.760312 valid_1's rmse: 1.34351
Early stopping, best iteration is:
[1271] training's rmse: 0.783589 valid_1's rmse: 1.34267
Model training done in 13.239583253860474 seconds.
Fold error 1.3426663735902713
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39605 valid_1's rmse: 1.48101
[300] training's rmse: 1.21257 valid_1's rmse: 1.3889
[450] training's rmse: 1.10051 valid_1's rmse: 1.35543
[600] training's rmse: 1.01902 valid_1's rmse: 1.3438
[750] training's rmse: 0.950629 valid_1's rmse: 1.33756
[900] training's rmse: 0.891626 valid_1's rmse: 1.33558
[1050] training's rmse: 0.839099 valid_1's rmse: 1.3338
[1200] training's rmse: 0.790876 valid_1's rmse: 1.33189
[1350] training's rmse: 0.746171 valid_1's rmse: 1.33115
[1500] training's rmse: 0.70686 valid_1's rmse: 1.33081
[1650] training's rmse: 0.66964 valid_1's rmse: 1.33017
[1800] training's rmse: 0.634485 valid_1's rmse: 1.33015
Early stopping, best iteration is:
[1730] training's rmse: 0.650656 valid_1's rmse: 1.32971
Model training done in 16.401126623153687 seconds.
Fold error 1.3297082087109375
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.38822 valid_1's rmse: 1.50809
[300] training's rmse: 1.20359 valid_1's rmse: 1.41228
[450] training's rmse: 1.09027 valid_1's rmse: 1.3785
[600] training's rmse: 1.00784 valid_1's rmse: 1.36723
[750] training's rmse: 0.938444 valid_1's rmse: 1.36194
[900] training's rmse: 0.878475 valid_1's rmse: 1.35843
[1050] training's rmse: 0.826543 valid_1's rmse: 1.35697
[1200] training's rmse: 0.778246 valid_1's rmse: 1.35651
[1350] training's rmse: 0.734462 valid_1's rmse: 1.35561
[1500] training's rmse: 0.694637 valid_1's rmse: 1.35532
[1650] training's rmse: 0.65599 valid_1's rmse: 1.35522
Early stopping, best iteration is:
[1659] training's rmse: 0.65367 valid_1's rmse: 1.35512
Model training done in 15.607249736785889 seconds.
Fold error 1.3551242746019705
Total error 1.329879070771735
Total std 0.02622
Fitting big fold 4 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.37563 valid_1's rmse: 1.4765
[300] training's rmse: 1.1693 valid_1's rmse: 1.37896
[450] training's rmse: 1.03662 valid_1's rmse: 1.34569
[600] training's rmse: 0.938664 valid_1's rmse: 1.33402
[750] training's rmse: 0.858466 valid_1's rmse: 1.32858
[900] training's rmse: 0.788512 valid_1's rmse: 1.32671
[1050] training's rmse: 0.727604 valid_1's rmse: 1.32513
[1200] training's rmse: 0.673194 valid_1's rmse: 1.32427
[1350] training's rmse: 0.624268 valid_1's rmse: 1.32404
[1500] training's rmse: 0.57828 valid_1's rmse: 1.32374
Early stopping, best iteration is:
[1495] training's rmse: 0.579816 valid_1's rmse: 1.32367
Model training done in 12.837659120559692 seconds.
Fold error 1.3236681967363249
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40351 valid_1's rmse: 1.46398
[300] training's rmse: 1.22699 valid_1's rmse: 1.36244
[450] training's rmse: 1.11951 valid_1's rmse: 1.32455
[600] training's rmse: 1.04163 valid_1's rmse: 1.31128
[750] training's rmse: 0.978448 valid_1's rmse: 1.30467
[900] training's rmse: 0.922759 valid_1's rmse: 1.30396
[1050] training's rmse: 0.873962 valid_1's rmse: 1.3028
[1200] training's rmse: 0.827899 valid_1's rmse: 1.30198
[1350] training's rmse: 0.784123 valid_1's rmse: 1.30071
[1500] training's rmse: 0.745127 valid_1's rmse: 1.30039
Early stopping, best iteration is:
[1521] training's rmse: 0.739505 valid_1's rmse: 1.30019
Model training done in 15.129538297653198 seconds.
Fold error 1.3001890459058187
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39636 valid_1's rmse: 1.50823
[300] training's rmse: 1.21823 valid_1's rmse: 1.40745
[450] training's rmse: 1.10985 valid_1's rmse: 1.3664
[600] training's rmse: 1.03087 valid_1's rmse: 1.34942
[750] training's rmse: 0.964789 valid_1's rmse: 1.33876
[900] training's rmse: 0.906931 valid_1's rmse: 1.33351
[1050] training's rmse: 0.855233 valid_1's rmse: 1.33003
[1200] training's rmse: 0.807195 valid_1's rmse: 1.32684
[1350] training's rmse: 0.762762 valid_1's rmse: 1.32517
[1500] training's rmse: 0.72239 valid_1's rmse: 1.32291
[1650] training's rmse: 0.683409 valid_1's rmse: 1.3215
[1800] training's rmse: 0.646413 valid_1's rmse: 1.32101
[1950] training's rmse: 0.612387 valid_1's rmse: 1.32016
[2100] training's rmse: 0.57993 valid_1's rmse: 1.31993
Early stopping, best iteration is:
[2065] training's rmse: 0.587453 valid_1's rmse: 1.31969
Model training done in 19.97556447982788 seconds.
Fold error 1.319693952855568
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39358 valid_1's rmse: 1.4917
[300] training's rmse: 1.21132 valid_1's rmse: 1.39829
[450] training's rmse: 1.10034 valid_1's rmse: 1.36763
[600] training's rmse: 1.01949 valid_1's rmse: 1.35478
[750] training's rmse: 0.951993 valid_1's rmse: 1.34947
[900] training's rmse: 0.892936 valid_1's rmse: 1.34658
[1050] training's rmse: 0.839422 valid_1's rmse: 1.3454
[1200] training's rmse: 0.78972 valid_1's rmse: 1.34354
[1350] training's rmse: 0.745942 valid_1's rmse: 1.34328
Early stopping, best iteration is:
[1309] training's rmse: 0.757827 valid_1's rmse: 1.34292
Model training done in 13.178747653961182 seconds.
Fold error 1.3429203913702492
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.38943 valid_1's rmse: 1.49472
[300] training's rmse: 1.20265 valid_1's rmse: 1.40565
[450] training's rmse: 1.08865 valid_1's rmse: 1.37334
[600] training's rmse: 1.006 valid_1's rmse: 1.362
[750] training's rmse: 0.937953 valid_1's rmse: 1.35602
[900] training's rmse: 0.877597 valid_1's rmse: 1.35294
[1050] training's rmse: 0.822751 valid_1's rmse: 1.35124
[1200] training's rmse: 0.773188 valid_1's rmse: 1.35041
Early stopping, best iteration is:
[1193] training's rmse: 0.775619 valid_1's rmse: 1.3503
Model training done in 12.609271049499512 seconds.
Fold error 1.3503034116641597
Total error 1.3293185634542535
Total std 0.01777
Fitting big fold 5 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.36876 valid_1's rmse: 1.49293
[300] training's rmse: 1.16288 valid_1's rmse: 1.4022
[450] training's rmse: 1.03233 valid_1's rmse: 1.37018
[600] training's rmse: 0.937095 valid_1's rmse: 1.35863
[750] training's rmse: 0.856618 valid_1's rmse: 1.35341
[900] training's rmse: 0.788286 valid_1's rmse: 1.35074
[1050] training's rmse: 0.728277 valid_1's rmse: 1.3494
Early stopping, best iteration is:
[1039] training's rmse: 0.732707 valid_1's rmse: 1.34929
Model training done in 9.381902933120728 seconds.
Fold error 1.3492915533629644
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40451 valid_1's rmse: 1.45457
[300] training's rmse: 1.22741 valid_1's rmse: 1.34433
[450] training's rmse: 1.11898 valid_1's rmse: 1.3018
[600] training's rmse: 1.04022 valid_1's rmse: 1.2863
[750] training's rmse: 0.975254 valid_1's rmse: 1.28033
[900] training's rmse: 0.918655 valid_1's rmse: 1.27721
[1050] training's rmse: 0.867344 valid_1's rmse: 1.27557
[1200] training's rmse: 0.820601 valid_1's rmse: 1.27447
Early stopping, best iteration is:
[1203] training's rmse: 0.819632 valid_1's rmse: 1.27444
Model training done in 12.486598253250122 seconds.
Fold error 1.2744380560382151
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40208 valid_1's rmse: 1.46425
[300] training's rmse: 1.22358 valid_1's rmse: 1.36429
[450] training's rmse: 1.11479 valid_1's rmse: 1.33228
[600] training's rmse: 1.03541 valid_1's rmse: 1.32111
[750] training's rmse: 0.969333 valid_1's rmse: 1.31645
[900] training's rmse: 0.912118 valid_1's rmse: 1.31442
[1050] training's rmse: 0.862341 valid_1's rmse: 1.31399
[1200] training's rmse: 0.814463 valid_1's rmse: 1.31316
[1350] training's rmse: 0.770996 valid_1's rmse: 1.31251
Early stopping, best iteration is:
[1356] training's rmse: 0.769301 valid_1's rmse: 1.31217
Model training done in 13.599620580673218 seconds.
Fold error 1.3121740603658691
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.38723 valid_1's rmse: 1.52267
[300] training's rmse: 1.20311 valid_1's rmse: 1.43843
[450] training's rmse: 1.0916 valid_1's rmse: 1.40927
[600] training's rmse: 1.01177 valid_1's rmse: 1.3982
[750] training's rmse: 0.944292 valid_1's rmse: 1.39424
[900] training's rmse: 0.884886 valid_1's rmse: 1.39256
[1050] training's rmse: 0.832306 valid_1's rmse: 1.39137
[1200] training's rmse: 0.783101 valid_1's rmse: 1.39056
Early stopping, best iteration is:
[1150] training's rmse: 0.799217 valid_1's rmse: 1.39049
Model training done in 11.862268447875977 seconds.
Fold error 1.390486478523729
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.3949 valid_1's rmse: 1.47556
[300] training's rmse: 1.20992 valid_1's rmse: 1.37377
[450] training's rmse: 1.09467 valid_1's rmse: 1.33885
[600] training's rmse: 1.0108 valid_1's rmse: 1.32576
[750] training's rmse: 0.941166 valid_1's rmse: 1.32034
[900] training's rmse: 0.879005 valid_1's rmse: 1.319
Early stopping, best iteration is:
[848] training's rmse: 0.89992 valid_1's rmse: 1.31854
Model training done in 8.979979991912842 seconds.
Fold error 1.318538984581108
Total error 1.3380344590037525
Total std 0.03889
Fitting big fold 6 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.37021 valid_1's rmse: 1.48485
[300] training's rmse: 1.16281 valid_1's rmse: 1.39498
[450] training's rmse: 1.03216 valid_1's rmse: 1.36417
[600] training's rmse: 0.936567 valid_1's rmse: 1.35095
[750] training's rmse: 0.857561 valid_1's rmse: 1.34488
[900] training's rmse: 0.788992 valid_1's rmse: 1.34062
[1050] training's rmse: 0.730832 valid_1's rmse: 1.339
[1200] training's rmse: 0.677221 valid_1's rmse: 1.3374
[1350] training's rmse: 0.628757 valid_1's rmse: 1.33624
[1500] training's rmse: 0.585785 valid_1's rmse: 1.33552
[1650] training's rmse: 0.544465 valid_1's rmse: 1.33509
[1800] training's rmse: 0.505584 valid_1's rmse: 1.33477
[1950] training's rmse: 0.470196 valid_1's rmse: 1.33443
[2100] training's rmse: 0.437494 valid_1's rmse: 1.33432
[2250] training's rmse: 0.407148 valid_1's rmse: 1.33418
Early stopping, best iteration is:
[2211] training's rmse: 0.414782 valid_1's rmse: 1.33403
Model training done in 18.206297397613525 seconds.
Fold error 1.3340260135554016
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40315 valid_1's rmse: 1.47184
[300] training's rmse: 1.22681 valid_1's rmse: 1.36392
[450] training's rmse: 1.11945 valid_1's rmse: 1.32541
[600] training's rmse: 1.04128 valid_1's rmse: 1.31117
[750] training's rmse: 0.976152 valid_1's rmse: 1.30498
[900] training's rmse: 0.919255 valid_1's rmse: 1.30106
[1050] training's rmse: 0.867364 valid_1's rmse: 1.29832
Early stopping, best iteration is:
[1052] training's rmse: 0.866686 valid_1's rmse: 1.29818
Model training done in 11.432419776916504 seconds.
Fold error 1.2981770929961984
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40236 valid_1's rmse: 1.46579
[300] training's rmse: 1.22438 valid_1's rmse: 1.35199
[450] training's rmse: 1.11505 valid_1's rmse: 1.30798
[600] training's rmse: 1.03553 valid_1's rmse: 1.29143
[750] training's rmse: 0.96922 valid_1's rmse: 1.28177
[900] training's rmse: 0.911716 valid_1's rmse: 1.27682
[1050] training's rmse: 0.859896 valid_1's rmse: 1.27373
[1200] training's rmse: 0.810767 valid_1's rmse: 1.27111
[1350] training's rmse: 0.767338 valid_1's rmse: 1.26938
[1500] training's rmse: 0.726865 valid_1's rmse: 1.26895
Early stopping, best iteration is:
[1453] training's rmse: 0.739176 valid_1's rmse: 1.26842
Model training done in 14.864237785339355 seconds.
Fold error 1.2684195954245827
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39738 valid_1's rmse: 1.46884
[300] training's rmse: 1.21551 valid_1's rmse: 1.37096
[450] training's rmse: 1.10449 valid_1's rmse: 1.33421
[600] training's rmse: 1.02366 valid_1's rmse: 1.31976
[750] training's rmse: 0.956852 valid_1's rmse: 1.31243
[900] training's rmse: 0.897871 valid_1's rmse: 1.30859
[1050] training's rmse: 0.843671 valid_1's rmse: 1.30686
[1200] training's rmse: 0.794691 valid_1's rmse: 1.30495
Early stopping, best iteration is:
[1234] training's rmse: 0.784331 valid_1's rmse: 1.30415
Model training done in 12.69204831123352 seconds.
Fold error 1.3041479659885422
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.38625 valid_1's rmse: 1.51354
[300] training's rmse: 1.20126 valid_1's rmse: 1.42112
[450] training's rmse: 1.0875 valid_1's rmse: 1.38917
[600] training's rmse: 1.00423 valid_1's rmse: 1.37892
[750] training's rmse: 0.935987 valid_1's rmse: 1.37365
[900] training's rmse: 0.875097 valid_1's rmse: 1.37187
[1050] training's rmse: 0.820345 valid_1's rmse: 1.37061
[1200] training's rmse: 0.769829 valid_1's rmse: 1.36964
[1350] training's rmse: 0.724201 valid_1's rmse: 1.36854
[1500] training's rmse: 0.682473 valid_1's rmse: 1.36864
Early stopping, best iteration is:
[1452] training's rmse: 0.695115 valid_1's rmse: 1.36809
Model training done in 14.432392835617065 seconds.
Fold error 1.3680874728910013
Total error 1.3240974058287944
Total std 0.03391
Fitting big fold 7 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.37392 valid_1's rmse: 1.47585
[300] training's rmse: 1.16791 valid_1's rmse: 1.38158
[450] training's rmse: 1.03678 valid_1's rmse: 1.34788
[600] training's rmse: 0.940775 valid_1's rmse: 1.33519
[750] training's rmse: 0.861165 valid_1's rmse: 1.32991
[900] training's rmse: 0.790906 valid_1's rmse: 1.32719
[1050] training's rmse: 0.729732 valid_1's rmse: 1.32499
[1200] training's rmse: 0.674522 valid_1's rmse: 1.32367
[1350] training's rmse: 0.625171 valid_1's rmse: 1.32262
[1500] training's rmse: 0.579936 valid_1's rmse: 1.32209
Early stopping, best iteration is:
[1515] training's rmse: 0.57563 valid_1's rmse: 1.32202
Model training done in 13.227648973464966 seconds.
Fold error 1.3220196656169263
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40052 valid_1's rmse: 1.49732
[300] training's rmse: 1.22357 valid_1's rmse: 1.39132
[450] training's rmse: 1.11622 valid_1's rmse: 1.35094
[600] training's rmse: 1.03797 valid_1's rmse: 1.33332
[750] training's rmse: 0.973087 valid_1's rmse: 1.32393
[900] training's rmse: 0.916341 valid_1's rmse: 1.31765
[1050] training's rmse: 0.865051 valid_1's rmse: 1.31565
[1200] training's rmse: 0.816873 valid_1's rmse: 1.31321
[1350] training's rmse: 0.772943 valid_1's rmse: 1.31028
[1500] training's rmse: 0.732509 valid_1's rmse: 1.30883
[1650] training's rmse: 0.694369 valid_1's rmse: 1.30777
[1800] training's rmse: 0.658939 valid_1's rmse: 1.30676
[1950] training's rmse: 0.624347 valid_1's rmse: 1.30604
[2100] training's rmse: 0.592403 valid_1's rmse: 1.30513
[2250] training's rmse: 0.562557 valid_1's rmse: 1.30496
[2400] training's rmse: 0.535055 valid_1's rmse: 1.30486
Early stopping, best iteration is:
[2301] training's rmse: 0.552825 valid_1's rmse: 1.3047
Model training done in 22.72022247314453 seconds.
Fold error 1.3047009301268246
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40482 valid_1's rmse: 1.43823
[300] training's rmse: 1.22598 valid_1's rmse: 1.33166
[450] training's rmse: 1.11611 valid_1's rmse: 1.2966
[600] training's rmse: 1.03724 valid_1's rmse: 1.28314
[750] training's rmse: 0.971176 valid_1's rmse: 1.27877
[900] training's rmse: 0.912435 valid_1's rmse: 1.27616
[1050] training's rmse: 0.85988 valid_1's rmse: 1.27511
Early stopping, best iteration is:
[1067] training's rmse: 0.853669 valid_1's rmse: 1.27477
Model training done in 11.535143375396729 seconds.
Fold error 1.2747698797716334
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39547 valid_1's rmse: 1.49635
[300] training's rmse: 1.21326 valid_1's rmse: 1.4017
[450] training's rmse: 1.10146 valid_1's rmse: 1.36808
[600] training's rmse: 1.01985 valid_1's rmse: 1.3547
[750] training's rmse: 0.952646 valid_1's rmse: 1.34765
[900] training's rmse: 0.893238 valid_1's rmse: 1.3442
[1050] training's rmse: 0.839547 valid_1's rmse: 1.34281
[1200] training's rmse: 0.791239 valid_1's rmse: 1.34129
[1350] training's rmse: 0.747469 valid_1's rmse: 1.34086
[1500] training's rmse: 0.706359 valid_1's rmse: 1.33979
[1650] training's rmse: 0.66722 valid_1's rmse: 1.33875
[1800] training's rmse: 0.631915 valid_1's rmse: 1.33807
[1950] training's rmse: 0.597461 valid_1's rmse: 1.33776
[2100] training's rmse: 0.564995 valid_1's rmse: 1.33755
Early stopping, best iteration is:
[2050] training's rmse: 0.575773 valid_1's rmse: 1.33737
Model training done in 20.11122751235962 seconds.
Fold error 1.3373666634068397
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.38747 valid_1's rmse: 1.51198
[300] training's rmse: 1.2017 valid_1's rmse: 1.41867
[450] training's rmse: 1.08705 valid_1's rmse: 1.38496
[600] training's rmse: 1.00406 valid_1's rmse: 1.37327
[750] training's rmse: 0.936504 valid_1's rmse: 1.36798
[900] training's rmse: 0.876738 valid_1's rmse: 1.36453
[1050] training's rmse: 0.823792 valid_1's rmse: 1.36196
[1200] training's rmse: 0.77464 valid_1's rmse: 1.36013
[1350] training's rmse: 0.730649 valid_1's rmse: 1.35851
[1500] training's rmse: 0.691359 valid_1's rmse: 1.35778
[1650] training's rmse: 0.65386 valid_1's rmse: 1.35714
[1800] training's rmse: 0.618917 valid_1's rmse: 1.35698
Early stopping, best iteration is:
[1772] training's rmse: 0.625905 valid_1's rmse: 1.35679
Model training done in 16.881840467453003 seconds.
Fold error 1.3567851284657952
Total error 1.3237838276642429
Total std 0.02805
Fitting big fold 8 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.36375 valid_1's rmse: 1.50354
[300] training's rmse: 1.15917 valid_1's rmse: 1.40499
[450] training's rmse: 1.02664 valid_1's rmse: 1.37124
[600] training's rmse: 0.930008 valid_1's rmse: 1.35726
[750] training's rmse: 0.850762 valid_1's rmse: 1.35131
[900] training's rmse: 0.780458 valid_1's rmse: 1.34811
[1050] training's rmse: 0.718593 valid_1's rmse: 1.34693
[1200] training's rmse: 0.661794 valid_1's rmse: 1.34635
Early stopping, best iteration is:
[1176] training's rmse: 0.67052 valid_1's rmse: 1.34609
Model training done in 10.500909566879272 seconds.
Fold error 1.3460859609935174
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.4025 valid_1's rmse: 1.47316
[300] training's rmse: 1.22451 valid_1's rmse: 1.37518
[450] training's rmse: 1.11589 valid_1's rmse: 1.33943
[600] training's rmse: 1.0379 valid_1's rmse: 1.32411
[750] training's rmse: 0.973221 valid_1's rmse: 1.3178
[900] training's rmse: 0.916253 valid_1's rmse: 1.31432
[1050] training's rmse: 0.865382 valid_1's rmse: 1.31151
[1200] training's rmse: 0.818759 valid_1's rmse: 1.31047
[1350] training's rmse: 0.776387 valid_1's rmse: 1.3101
[1500] training's rmse: 0.736781 valid_1's rmse: 1.30948
[1650] training's rmse: 0.698049 valid_1's rmse: 1.30821
[1800] training's rmse: 0.662251 valid_1's rmse: 1.30817
Early stopping, best iteration is:
[1731] training's rmse: 0.679196 valid_1's rmse: 1.30777
Model training done in 17.64180898666382 seconds.
Fold error 1.3077704984926803
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.4028 valid_1's rmse: 1.44499
[300] training's rmse: 1.22203 valid_1's rmse: 1.35309
[450] training's rmse: 1.11219 valid_1's rmse: 1.32241
[600] training's rmse: 1.0329 valid_1's rmse: 1.31255
[750] training's rmse: 0.966594 valid_1's rmse: 1.30664
[900] training's rmse: 0.908913 valid_1's rmse: 1.30387
[1050] training's rmse: 0.856891 valid_1's rmse: 1.30266
[1200] training's rmse: 0.809058 valid_1's rmse: 1.30254
[1350] training's rmse: 0.764591 valid_1's rmse: 1.30108
[1500] training's rmse: 0.724469 valid_1's rmse: 1.30093
Early stopping, best iteration is:
[1407] training's rmse: 0.749162 valid_1's rmse: 1.30088
Model training done in 15.014834880828857 seconds.
Fold error 1.300883369347001
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39531 valid_1's rmse: 1.49272
[300] training's rmse: 1.21396 valid_1's rmse: 1.3973
[450] training's rmse: 1.10434 valid_1's rmse: 1.36399
[600] training's rmse: 1.02437 valid_1's rmse: 1.35287
[750] training's rmse: 0.957792 valid_1's rmse: 1.34677
[900] training's rmse: 0.898888 valid_1's rmse: 1.34414
[1050] training's rmse: 0.846426 valid_1's rmse: 1.3432
Early stopping, best iteration is:
[989] training's rmse: 0.866854 valid_1's rmse: 1.34302
Model training done in 11.41247296333313 seconds.
Fold error 1.3430214430755516
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39274 valid_1's rmse: 1.48023
[300] training's rmse: 1.20797 valid_1's rmse: 1.38108
[450] training's rmse: 1.09443 valid_1's rmse: 1.34586
[600] training's rmse: 1.01142 valid_1's rmse: 1.33352
[750] training's rmse: 0.94271 valid_1's rmse: 1.3262
[900] training's rmse: 0.882366 valid_1's rmse: 1.32308
[1050] training's rmse: 0.828167 valid_1's rmse: 1.31985
[1200] training's rmse: 0.778281 valid_1's rmse: 1.31859
[1350] training's rmse: 0.732687 valid_1's rmse: 1.31721
[1500] training's rmse: 0.691095 valid_1's rmse: 1.31607
[1650] training's rmse: 0.65155 valid_1's rmse: 1.31481
[1800] training's rmse: 0.614615 valid_1's rmse: 1.3144
Early stopping, best iteration is:
[1781] training's rmse: 0.619062 valid_1's rmse: 1.31428
Model training done in 17.400453329086304 seconds.
Fold error 1.3142804099985645
Total error 1.3295313930748704
Total std 0.0186
Fitting big fold 9 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.36575 valid_1's rmse: 1.49502
[300] training's rmse: 1.15759 valid_1's rmse: 1.40672
[450] training's rmse: 1.0251 valid_1's rmse: 1.37598
[600] training's rmse: 0.927601 valid_1's rmse: 1.36571
[750] training's rmse: 0.846726 valid_1's rmse: 1.3614
[900] training's rmse: 0.775496 valid_1's rmse: 1.36031
[1050] training's rmse: 0.714415 valid_1's rmse: 1.35897
[1200] training's rmse: 0.657344 valid_1's rmse: 1.35847
[1350] training's rmse: 0.607335 valid_1's rmse: 1.3583
Early stopping, best iteration is:
[1375] training's rmse: 0.599166 valid_1's rmse: 1.35816
Model training done in 12.388858795166016 seconds.
Fold error 1.3581625362698362
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.4016 valid_1's rmse: 1.47426
[300] training's rmse: 1.22347 valid_1's rmse: 1.38296
[450] training's rmse: 1.11502 valid_1's rmse: 1.35075
[600] training's rmse: 1.03748 valid_1's rmse: 1.33823
[750] training's rmse: 0.972843 valid_1's rmse: 1.3322
[900] training's rmse: 0.916559 valid_1's rmse: 1.32964
[1050] training's rmse: 0.865763 valid_1's rmse: 1.32931
[1200] training's rmse: 0.819 valid_1's rmse: 1.32815
[1350] training's rmse: 0.776923 valid_1's rmse: 1.32701
[1500] training's rmse: 0.738707 valid_1's rmse: 1.32542
[1650] training's rmse: 0.700697 valid_1's rmse: 1.32455
[1800] training's rmse: 0.666257 valid_1's rmse: 1.32324
[1950] training's rmse: 0.631733 valid_1's rmse: 1.32272
Early stopping, best iteration is:
[1997] training's rmse: 0.621861 valid_1's rmse: 1.32226
Model training done in 19.702285051345825 seconds.
Fold error 1.32225746642199
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40024 valid_1's rmse: 1.47106
[300] training's rmse: 1.22071 valid_1's rmse: 1.37447
[450] training's rmse: 1.11026 valid_1's rmse: 1.33897
[600] training's rmse: 1.03119 valid_1's rmse: 1.32326
[750] training's rmse: 0.964884 valid_1's rmse: 1.31613
[900] training's rmse: 0.90557 valid_1's rmse: 1.31127
[1050] training's rmse: 0.852963 valid_1's rmse: 1.30844
[1200] training's rmse: 0.80339 valid_1's rmse: 1.30776
[1350] training's rmse: 0.758317 valid_1's rmse: 1.30667
[1500] training's rmse: 0.71801 valid_1's rmse: 1.30488
Early stopping, best iteration is:
[1496] training's rmse: 0.719103 valid_1's rmse: 1.30476
Model training done in 15.775087594985962 seconds.
Fold error 1.3047576398269047
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.38996 valid_1's rmse: 1.51522
[300] training's rmse: 1.20849 valid_1's rmse: 1.41626
[450] training's rmse: 1.09789 valid_1's rmse: 1.38257
[600] training's rmse: 1.01711 valid_1's rmse: 1.36904
[750] training's rmse: 0.951169 valid_1's rmse: 1.36258
[900] training's rmse: 0.89257 valid_1's rmse: 1.3583
[1050] training's rmse: 0.841558 valid_1's rmse: 1.35602
[1200] training's rmse: 0.792838 valid_1's rmse: 1.35449
[1350] training's rmse: 0.749176 valid_1's rmse: 1.35375
[1500] training's rmse: 0.708964 valid_1's rmse: 1.35405
Early stopping, best iteration is:
[1468] training's rmse: 0.717889 valid_1's rmse: 1.35348
Model training done in 15.172836065292358 seconds.
Fold error 1.3534832769014924
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40014 valid_1's rmse: 1.45665
[300] training's rmse: 1.21671 valid_1's rmse: 1.34756
[450] training's rmse: 1.1029 valid_1's rmse: 1.30873
[600] training's rmse: 1.01898 valid_1's rmse: 1.29395
[750] training's rmse: 0.949838 valid_1's rmse: 1.28803
[900] training's rmse: 0.889543 valid_1's rmse: 1.28612
[1050] training's rmse: 0.835088 valid_1's rmse: 1.28531
[1200] training's rmse: 0.784042 valid_1's rmse: 1.28414
Early stopping, best iteration is:
[1208] training's rmse: 0.781532 valid_1's rmse: 1.28394
Model training done in 12.705013751983643 seconds.
Fold error 1.2839416319002654
Total error 1.3323204429989834
Total std 0.02833
Fitting big fold 10 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.37147 valid_1's rmse: 1.49136
[300] training's rmse: 1.16823 valid_1's rmse: 1.39024
[450] training's rmse: 1.03784 valid_1's rmse: 1.35298
[600] training's rmse: 0.940552 valid_1's rmse: 1.33813
[750] training's rmse: 0.860722 valid_1's rmse: 1.33098
[900] training's rmse: 0.793019 valid_1's rmse: 1.32748
[1050] training's rmse: 0.73399 valid_1's rmse: 1.32541
[1200] training's rmse: 0.677701 valid_1's rmse: 1.32333
[1350] training's rmse: 0.627893 valid_1's rmse: 1.32227
[1500] training's rmse: 0.583052 valid_1's rmse: 1.32158
[1650] training's rmse: 0.540189 valid_1's rmse: 1.32134
[1800] training's rmse: 0.502307 valid_1's rmse: 1.32112
Early stopping, best iteration is:
[1708] training's rmse: 0.525 valid_1's rmse: 1.32108
Model training done in 14.879196166992188 seconds.
Fold error 1.3210773038965957
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.41011 valid_1's rmse: 1.39308
[300] training's rmse: 1.23256 valid_1's rmse: 1.28547
[450] training's rmse: 1.12414 valid_1's rmse: 1.24678
[600] training's rmse: 1.04533 valid_1's rmse: 1.23094
[750] training's rmse: 0.980185 valid_1's rmse: 1.22453
[900] training's rmse: 0.923903 valid_1's rmse: 1.22119
[1050] training's rmse: 0.872106 valid_1's rmse: 1.22005
[1200] training's rmse: 0.824229 valid_1's rmse: 1.2171
[1350] training's rmse: 0.780572 valid_1's rmse: 1.21669
Early stopping, best iteration is:
[1259] training's rmse: 0.806305 valid_1's rmse: 1.21656
Model training done in 13.163786172866821 seconds.
Fold error 1.2165560342511965
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40287 valid_1's rmse: 1.45134
[300] training's rmse: 1.22326 valid_1's rmse: 1.35217
[450] training's rmse: 1.11422 valid_1's rmse: 1.31697
[600] training's rmse: 1.03469 valid_1's rmse: 1.30418
[750] training's rmse: 0.969545 valid_1's rmse: 1.29632
[900] training's rmse: 0.911737 valid_1's rmse: 1.29135
[1050] training's rmse: 0.858198 valid_1's rmse: 1.28945
[1200] training's rmse: 0.811335 valid_1's rmse: 1.28776
[1350] training's rmse: 0.767972 valid_1's rmse: 1.28714
Early stopping, best iteration is:
[1255] training's rmse: 0.794711 valid_1's rmse: 1.28665
Model training done in 13.08300256729126 seconds.
Fold error 1.2866452191998947
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39135 valid_1's rmse: 1.50024
[300] training's rmse: 1.20768 valid_1's rmse: 1.41455
[450] training's rmse: 1.09567 valid_1's rmse: 1.38326
[600] training's rmse: 1.01463 valid_1's rmse: 1.37119
[750] training's rmse: 0.946979 valid_1's rmse: 1.3646
[900] training's rmse: 0.888312 valid_1's rmse: 1.3609
[1050] training's rmse: 0.835491 valid_1's rmse: 1.35905
[1200] training's rmse: 0.786244 valid_1's rmse: 1.358
[1350] training's rmse: 0.742849 valid_1's rmse: 1.35765
Early stopping, best iteration is:
[1389] training's rmse: 0.731354 valid_1's rmse: 1.35737
Model training done in 13.991573810577393 seconds.
Fold error 1.3573671934879719
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.38388 valid_1's rmse: 1.52573
[300] training's rmse: 1.19914 valid_1's rmse: 1.42913
[450] training's rmse: 1.08499 valid_1's rmse: 1.39592
[600] training's rmse: 1.00148 valid_1's rmse: 1.38277
[750] training's rmse: 0.932549 valid_1's rmse: 1.37688
[900] training's rmse: 0.873165 valid_1's rmse: 1.37409
[1050] training's rmse: 0.820457 valid_1's rmse: 1.37172
[1200] training's rmse: 0.770823 valid_1's rmse: 1.36995
[1350] training's rmse: 0.726154 valid_1's rmse: 1.36849
[1500] training's rmse: 0.684575 valid_1's rmse: 1.36801
[1650] training's rmse: 0.644357 valid_1's rmse: 1.36711
Early stopping, best iteration is:
[1645] training's rmse: 0.645612 valid_1's rmse: 1.36702
Model training done in 15.933378219604492 seconds.
Fold error 1.3670188238686782
Total error 1.3220313559644286
Total std 0.05458
Fitting big fold 11 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.364 valid_1's rmse: 1.50687
[300] training's rmse: 1.15982 valid_1's rmse: 1.41031
[450] training's rmse: 1.0325 valid_1's rmse: 1.37558
[600] training's rmse: 0.937586 valid_1's rmse: 1.36219
[750] training's rmse: 0.859899 valid_1's rmse: 1.3556
[900] training's rmse: 0.793115 valid_1's rmse: 1.35305
[1050] training's rmse: 0.733601 valid_1's rmse: 1.35164
[1200] training's rmse: 0.6785 valid_1's rmse: 1.351
[1350] training's rmse: 0.628209 valid_1's rmse: 1.35106
[1500] training's rmse: 0.583294 valid_1's rmse: 1.35036
Early stopping, best iteration is:
[1529] training's rmse: 0.574597 valid_1's rmse: 1.35012
Model training done in 13.082005500793457 seconds.
Fold error 1.3501154301810236
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40633 valid_1's rmse: 1.40435
[300] training's rmse: 1.22638 valid_1's rmse: 1.31979
[450] training's rmse: 1.11703 valid_1's rmse: 1.2951
[600] training's rmse: 1.03912 valid_1's rmse: 1.28778
[750] training's rmse: 0.973104 valid_1's rmse: 1.28482
[900] training's rmse: 0.916049 valid_1's rmse: 1.28256
[1050] training's rmse: 0.864742 valid_1's rmse: 1.28103
[1200] training's rmse: 0.817276 valid_1's rmse: 1.28018
[1350] training's rmse: 0.772937 valid_1's rmse: 1.27975
Early stopping, best iteration is:
[1338] training's rmse: 0.776109 valid_1's rmse: 1.2795
Model training done in 14.00254225730896 seconds.
Fold error 1.2794999589870928
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40198 valid_1's rmse: 1.4721
[300] training's rmse: 1.22442 valid_1's rmse: 1.36266
[450] training's rmse: 1.11569 valid_1's rmse: 1.32186
[600] training's rmse: 1.03689 valid_1's rmse: 1.30423
[750] training's rmse: 0.971733 valid_1's rmse: 1.29773
[900] training's rmse: 0.914103 valid_1's rmse: 1.2926
[1050] training's rmse: 0.862078 valid_1's rmse: 1.28918
[1200] training's rmse: 0.813904 valid_1's rmse: 1.28647
[1350] training's rmse: 0.7714 valid_1's rmse: 1.28522
Early stopping, best iteration is:
[1330] training's rmse: 0.776967 valid_1's rmse: 1.28479
Model training done in 13.643502712249756 seconds.
Fold error 1.2847898834520666
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39812 valid_1's rmse: 1.46845
[300] training's rmse: 1.21607 valid_1's rmse: 1.36728
[450] training's rmse: 1.10476 valid_1's rmse: 1.33199
[600] training's rmse: 1.02295 valid_1's rmse: 1.32026
[750] training's rmse: 0.955774 valid_1's rmse: 1.31494
[900] training's rmse: 0.897204 valid_1's rmse: 1.31353
[1050] training's rmse: 0.844253 valid_1's rmse: 1.31248
[1200] training's rmse: 0.795795 valid_1's rmse: 1.31152
Early stopping, best iteration is:
[1211] training's rmse: 0.7922 valid_1's rmse: 1.31135
Model training done in 12.387861490249634 seconds.
Fold error 1.3113524949289657
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.38705 valid_1's rmse: 1.51097
[300] training's rmse: 1.19981 valid_1's rmse: 1.4216
[450] training's rmse: 1.08456 valid_1's rmse: 1.39118
[600] training's rmse: 1.00128 valid_1's rmse: 1.37982
[750] training's rmse: 0.931293 valid_1's rmse: 1.3749
[900] training's rmse: 0.871507 valid_1's rmse: 1.3725
[1050] training's rmse: 0.816985 valid_1's rmse: 1.3709
[1200] training's rmse: 0.767634 valid_1's rmse: 1.37022
Early stopping, best iteration is:
[1108] training's rmse: 0.797551 valid_1's rmse: 1.36993
Model training done in 11.389532327651978 seconds.
Fold error 1.369932050972782
Total error 1.3323399161182061
Total std 0.03564
Fitting big fold 12 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.38124 valid_1's rmse: 1.47723
[300] training's rmse: 1.17634 valid_1's rmse: 1.37805
[450] training's rmse: 1.04556 valid_1's rmse: 1.3421
[600] training's rmse: 0.948626 valid_1's rmse: 1.32755
[750] training's rmse: 0.866875 valid_1's rmse: 1.31965
[900] training's rmse: 0.798584 valid_1's rmse: 1.31515
[1050] training's rmse: 0.737802 valid_1's rmse: 1.31328
[1200] training's rmse: 0.683854 valid_1's rmse: 1.31211
[1350] training's rmse: 0.633314 valid_1's rmse: 1.31172
[1500] training's rmse: 0.587084 valid_1's rmse: 1.31132
[1650] training's rmse: 0.544867 valid_1's rmse: 1.31155
Early stopping, best iteration is:
[1603] training's rmse: 0.557871 valid_1's rmse: 1.31111
Model training done in 13.692373752593994 seconds.
Fold error 1.3111116271086178
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.3975 valid_1's rmse: 1.50235
[300] training's rmse: 1.21963 valid_1's rmse: 1.40255
[450] training's rmse: 1.11155 valid_1's rmse: 1.36631
[600] training's rmse: 1.03317 valid_1's rmse: 1.35264
[750] training's rmse: 0.968914 valid_1's rmse: 1.34655
[900] training's rmse: 0.912367 valid_1's rmse: 1.34567
[1050] training's rmse: 0.862014 valid_1's rmse: 1.34451
[1200] training's rmse: 0.815067 valid_1's rmse: 1.34259
Early stopping, best iteration is:
[1190] training's rmse: 0.818212 valid_1's rmse: 1.3423
Model training done in 13.892840385437012 seconds.
Fold error 1.3423036205971588
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40048 valid_1's rmse: 1.47164
[300] training's rmse: 1.22102 valid_1's rmse: 1.378
[450] training's rmse: 1.11112 valid_1's rmse: 1.34544
[600] training's rmse: 1.03198 valid_1's rmse: 1.33283
[750] training's rmse: 0.96555 valid_1's rmse: 1.32622
[900] training's rmse: 0.9079 valid_1's rmse: 1.32439
[1050] training's rmse: 0.856207 valid_1's rmse: 1.32391
Early stopping, best iteration is:
[983] training's rmse: 0.87833 valid_1's rmse: 1.32361
Model training done in 13.436012744903564 seconds.
Fold error 1.3236096561028483
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39806 valid_1's rmse: 1.47462
[300] training's rmse: 1.21537 valid_1's rmse: 1.37787
[450] training's rmse: 1.10359 valid_1's rmse: 1.34597
[600] training's rmse: 1.02245 valid_1's rmse: 1.33496
[750] training's rmse: 0.955546 valid_1's rmse: 1.32936
[900] training's rmse: 0.896852 valid_1's rmse: 1.32616
[1050] training's rmse: 0.842649 valid_1's rmse: 1.3239
[1200] training's rmse: 0.792929 valid_1's rmse: 1.32176
[1350] training's rmse: 0.747438 valid_1's rmse: 1.32054
Early stopping, best iteration is:
[1370] training's rmse: 0.741864 valid_1's rmse: 1.32018
Model training done in 17.173158407211304 seconds.
Fold error 1.3201782741913137
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.38572 valid_1's rmse: 1.51154
[300] training's rmse: 1.20053 valid_1's rmse: 1.4153
[450] training's rmse: 1.08607 valid_1's rmse: 1.38226
[600] training's rmse: 1.00365 valid_1's rmse: 1.36817
[750] training's rmse: 0.935655 valid_1's rmse: 1.36216
[900] training's rmse: 0.876412 valid_1's rmse: 1.35821
[1050] training's rmse: 0.821796 valid_1's rmse: 1.3563
[1200] training's rmse: 0.774155 valid_1's rmse: 1.35353
[1350] training's rmse: 0.72901 valid_1's rmse: 1.35182
[1500] training's rmse: 0.688674 valid_1's rmse: 1.35092
[1650] training's rmse: 0.649925 valid_1's rmse: 1.35031
[1800] training's rmse: 0.614091 valid_1's rmse: 1.34944
[1950] training's rmse: 0.581717 valid_1's rmse: 1.34896
[2100] training's rmse: 0.54905 valid_1's rmse: 1.34843
[2250] training's rmse: 0.519015 valid_1's rmse: 1.34828
Early stopping, best iteration is:
[2198] training's rmse: 0.528977 valid_1's rmse: 1.34806
Model training done in 20.06234121322632 seconds.
Fold error 1.348057946304623
Total error 1.3247837470685224
Total std 0.01391
Fitting big fold 13 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.36707 valid_1's rmse: 1.49396
[300] training's rmse: 1.15823 valid_1's rmse: 1.40193
[450] training's rmse: 1.02519 valid_1's rmse: 1.36903
[600] training's rmse: 0.92678 valid_1's rmse: 1.35676
[750] training's rmse: 0.84687 valid_1's rmse: 1.35081
[900] training's rmse: 0.77787 valid_1's rmse: 1.34806
[1050] training's rmse: 0.716789 valid_1's rmse: 1.34687
[1200] training's rmse: 0.661132 valid_1's rmse: 1.34618
[1350] training's rmse: 0.611131 valid_1's rmse: 1.346
Early stopping, best iteration is:
[1341] training's rmse: 0.614085 valid_1's rmse: 1.34585
Model training done in 11.50821852684021 seconds.
Fold error 1.345851380629132
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40244 valid_1's rmse: 1.46341
[300] training's rmse: 1.22453 valid_1's rmse: 1.36573
[450] training's rmse: 1.11619 valid_1's rmse: 1.33378
[600] training's rmse: 1.03779 valid_1's rmse: 1.32159
[750] training's rmse: 0.972317 valid_1's rmse: 1.31716
[900] training's rmse: 0.915351 valid_1's rmse: 1.31568
Early stopping, best iteration is:
[936] training's rmse: 0.902382 valid_1's rmse: 1.31522
Model training done in 10.084029197692871 seconds.
Fold error 1.3152179834438404
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39972 valid_1's rmse: 1.4641
[300] training's rmse: 1.21916 valid_1's rmse: 1.37234
[450] training's rmse: 1.10981 valid_1's rmse: 1.34261
[600] training's rmse: 1.02909 valid_1's rmse: 1.33433
[750] training's rmse: 0.962863 valid_1's rmse: 1.33016
[900] training's rmse: 0.903847 valid_1's rmse: 1.32801
[1050] training's rmse: 0.851555 valid_1's rmse: 1.326
[1200] training's rmse: 0.803158 valid_1's rmse: 1.32572
[1350] training's rmse: 0.759485 valid_1's rmse: 1.32596
Early stopping, best iteration is:
[1270] training's rmse: 0.782627 valid_1's rmse: 1.32526
Model training done in 12.811732292175293 seconds.
Fold error 1.3252610919102428
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39285 valid_1's rmse: 1.50931
[300] training's rmse: 1.21323 valid_1's rmse: 1.40348
[450] training's rmse: 1.10296 valid_1's rmse: 1.36486
[600] training's rmse: 1.02303 valid_1's rmse: 1.34878
[750] training's rmse: 0.95666 valid_1's rmse: 1.34035
[900] training's rmse: 0.898361 valid_1's rmse: 1.33435
[1050] training's rmse: 0.845257 valid_1's rmse: 1.33094
[1200] training's rmse: 0.796242 valid_1's rmse: 1.32815
[1350] training's rmse: 0.752543 valid_1's rmse: 1.32644
[1500] training's rmse: 0.711278 valid_1's rmse: 1.32538
[1650] training's rmse: 0.671863 valid_1's rmse: 1.32473
[1800] training's rmse: 0.63599 valid_1's rmse: 1.32395
[1950] training's rmse: 0.602191 valid_1's rmse: 1.32329
Early stopping, best iteration is:
[1883] training's rmse: 0.616974 valid_1's rmse: 1.32299
Model training done in 17.75750184059143 seconds.
Fold error 1.3229916862131161
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39561 valid_1's rmse: 1.47946
[300] training's rmse: 1.21131 valid_1's rmse: 1.3803
[450] training's rmse: 1.09774 valid_1's rmse: 1.34421
[600] training's rmse: 1.01547 valid_1's rmse: 1.3313
[750] training's rmse: 0.947489 valid_1's rmse: 1.32439
[900] training's rmse: 0.88806 valid_1's rmse: 1.32073
[1050] training's rmse: 0.833781 valid_1's rmse: 1.31892
[1200] training's rmse: 0.783835 valid_1's rmse: 1.31753
[1350] training's rmse: 0.737248 valid_1's rmse: 1.3168
[1500] training's rmse: 0.695053 valid_1's rmse: 1.31659
[1650] training's rmse: 0.655244 valid_1's rmse: 1.31617
[1800] training's rmse: 0.618901 valid_1's rmse: 1.31546
[1950] training's rmse: 0.584842 valid_1's rmse: 1.31488
[2100] training's rmse: 0.552119 valid_1's rmse: 1.31449
[2250] training's rmse: 0.522548 valid_1's rmse: 1.3148
Early stopping, best iteration is:
[2154] training's rmse: 0.540991 valid_1's rmse: 1.3144
Model training done in 19.825969696044922 seconds.
Fold error 1.3143983316738788
Total error 1.3303006839685805
Total std 0.01137
Fitting big fold 14 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.37394 valid_1's rmse: 1.48164
[300] training's rmse: 1.16708 valid_1's rmse: 1.38966
[450] training's rmse: 1.03386 valid_1's rmse: 1.3566
[600] training's rmse: 0.936019 valid_1's rmse: 1.34441
[750] training's rmse: 0.854537 valid_1's rmse: 1.33864
[900] training's rmse: 0.784409 valid_1's rmse: 1.33573
[1050] training's rmse: 0.724259 valid_1's rmse: 1.33384
[1200] training's rmse: 0.669188 valid_1's rmse: 1.33338
Early stopping, best iteration is:
[1172] training's rmse: 0.678888 valid_1's rmse: 1.33327
Model training done in 11.448378801345825 seconds.
Fold error 1.3332670329293155
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40103 valid_1's rmse: 1.46995
[300] training's rmse: 1.22309 valid_1's rmse: 1.37944
[450] training's rmse: 1.11527 valid_1's rmse: 1.34756
[600] training's rmse: 1.03781 valid_1's rmse: 1.33433
[750] training's rmse: 0.973372 valid_1's rmse: 1.32994
[900] training's rmse: 0.916467 valid_1's rmse: 1.32853
[1050] training's rmse: 0.866745 valid_1's rmse: 1.32722
Early stopping, best iteration is:
[1086] training's rmse: 0.855163 valid_1's rmse: 1.32649
Model training done in 12.248238563537598 seconds.
Fold error 1.3264895542164825
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39556 valid_1's rmse: 1.49661
[300] training's rmse: 1.21559 valid_1's rmse: 1.40373
[450] training's rmse: 1.10488 valid_1's rmse: 1.37231
[600] training's rmse: 1.02429 valid_1's rmse: 1.36416
[750] training's rmse: 0.958012 valid_1's rmse: 1.3612
[900] training's rmse: 0.899562 valid_1's rmse: 1.36113
Early stopping, best iteration is:
[855] training's rmse: 0.916717 valid_1's rmse: 1.36026
Model training done in 11.10640263557434 seconds.
Fold error 1.3602560345397972
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39639 valid_1's rmse: 1.49405
[300] training's rmse: 1.2152 valid_1's rmse: 1.3932
[450] training's rmse: 1.10387 valid_1's rmse: 1.35588
[600] training's rmse: 1.02323 valid_1's rmse: 1.33947
[750] training's rmse: 0.956005 valid_1's rmse: 1.33085
[900] training's rmse: 0.897823 valid_1's rmse: 1.32612
[1050] training's rmse: 0.845565 valid_1's rmse: 1.32074
[1200] training's rmse: 0.797997 valid_1's rmse: 1.31698
[1350] training's rmse: 0.755176 valid_1's rmse: 1.31535
[1500] training's rmse: 0.714264 valid_1's rmse: 1.31393
[1650] training's rmse: 0.676061 valid_1's rmse: 1.31262
[1800] training's rmse: 0.639856 valid_1's rmse: 1.31103
[1950] training's rmse: 0.606103 valid_1's rmse: 1.30982
[2100] training's rmse: 0.5744 valid_1's rmse: 1.309
[2250] training's rmse: 0.544615 valid_1's rmse: 1.30813
[2400] training's rmse: 0.516265 valid_1's rmse: 1.30765
Early stopping, best iteration is:
[2407] training's rmse: 0.514661 valid_1's rmse: 1.30759
Model training done in 25.992780685424805 seconds.
Fold error 1.307585246373314
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39319 valid_1's rmse: 1.48997
[300] training's rmse: 1.20932 valid_1's rmse: 1.39064
[450] training's rmse: 1.09583 valid_1's rmse: 1.35377
[600] training's rmse: 1.01327 valid_1's rmse: 1.33963
[750] training's rmse: 0.944583 valid_1's rmse: 1.33149
[900] training's rmse: 0.884664 valid_1's rmse: 1.32696
[1050] training's rmse: 0.832736 valid_1's rmse: 1.32477
[1200] training's rmse: 0.784111 valid_1's rmse: 1.32299
[1350] training's rmse: 0.739319 valid_1's rmse: 1.32061
[1500] training's rmse: 0.698702 valid_1's rmse: 1.31897
[1650] training's rmse: 0.659531 valid_1's rmse: 1.31824
[1800] training's rmse: 0.623401 valid_1's rmse: 1.31741
[1950] training's rmse: 0.589344 valid_1's rmse: 1.31679
[2100] training's rmse: 0.557277 valid_1's rmse: 1.31566
[2250] training's rmse: 0.526967 valid_1's rmse: 1.31501
[2400] training's rmse: 0.498321 valid_1's rmse: 1.31467
[2550] training's rmse: 0.47204 valid_1's rmse: 1.31458
Early stopping, best iteration is:
[2505] training's rmse: 0.479775 valid_1's rmse: 1.3145
Model training done in 24.71419644355774 seconds.
Fold error 1.3145003389057641
Total error 1.3283620535946883
Total std 0.01827
Fitting big fold 15 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.36698 valid_1's rmse: 1.50135
[300] training's rmse: 1.16445 valid_1's rmse: 1.40189
[450] training's rmse: 1.03274 valid_1's rmse: 1.36623
[600] training's rmse: 0.936486 valid_1's rmse: 1.35222
[750] training's rmse: 0.856681 valid_1's rmse: 1.34426
[900] training's rmse: 0.787355 valid_1's rmse: 1.34044
[1050] training's rmse: 0.725486 valid_1's rmse: 1.33834
[1200] training's rmse: 0.670402 valid_1's rmse: 1.33689
[1350] training's rmse: 0.619997 valid_1's rmse: 1.33607
[1500] training's rmse: 0.575387 valid_1's rmse: 1.33561
[1650] training's rmse: 0.533068 valid_1's rmse: 1.33537
[1800] training's rmse: 0.495214 valid_1's rmse: 1.33495
Early stopping, best iteration is:
[1733] training's rmse: 0.51174 valid_1's rmse: 1.3348
Model training done in 15.811070203781128 seconds.
Fold error 1.334804357285626
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40021 valid_1's rmse: 1.48419
[300] training's rmse: 1.22168 valid_1's rmse: 1.39793
[450] training's rmse: 1.11438 valid_1's rmse: 1.36647
[600] training's rmse: 1.03672 valid_1's rmse: 1.35329
[750] training's rmse: 0.972708 valid_1's rmse: 1.3447
[900] training's rmse: 0.917625 valid_1's rmse: 1.33994
[1050] training's rmse: 0.868019 valid_1's rmse: 1.33681
[1200] training's rmse: 0.820583 valid_1's rmse: 1.33553
[1350] training's rmse: 0.777909 valid_1's rmse: 1.33201
[1500] training's rmse: 0.738336 valid_1's rmse: 1.33017
[1650] training's rmse: 0.700268 valid_1's rmse: 1.32855
[1800] training's rmse: 0.664325 valid_1's rmse: 1.32706
[1950] training's rmse: 0.630372 valid_1's rmse: 1.32533
[2100] training's rmse: 0.598242 valid_1's rmse: 1.32525
[2250] training's rmse: 0.568023 valid_1's rmse: 1.32457
[2400] training's rmse: 0.540276 valid_1's rmse: 1.32386
[2550] training's rmse: 0.514735 valid_1's rmse: 1.32391
[2700] training's rmse: 0.489551 valid_1's rmse: 1.32366
Early stopping, best iteration is:
[2682] training's rmse: 0.492428 valid_1's rmse: 1.32333
Model training done in 27.483307600021362 seconds.
Fold error 1.3233337562298855
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39747 valid_1's rmse: 1.4906
[300] training's rmse: 1.21862 valid_1's rmse: 1.38446
[450] training's rmse: 1.10893 valid_1's rmse: 1.34824
[600] training's rmse: 1.02977 valid_1's rmse: 1.33434
[750] training's rmse: 0.963112 valid_1's rmse: 1.3267
[900] training's rmse: 0.906079 valid_1's rmse: 1.32326
[1050] training's rmse: 0.854084 valid_1's rmse: 1.32102
[1200] training's rmse: 0.805235 valid_1's rmse: 1.3204
[1350] training's rmse: 0.761813 valid_1's rmse: 1.3194
Early stopping, best iteration is:
[1367] training's rmse: 0.756928 valid_1's rmse: 1.31909
Model training done in 14.88203477859497 seconds.
Fold error 1.319088647790788
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40099 valid_1's rmse: 1.45142
[300] training's rmse: 1.21803 valid_1's rmse: 1.35766
[450] training's rmse: 1.10639 valid_1's rmse: 1.32426
[600] training's rmse: 1.02509 valid_1's rmse: 1.31339
[750] training's rmse: 0.957346 valid_1's rmse: 1.30777
[900] training's rmse: 0.898539 valid_1's rmse: 1.30505
[1050] training's rmse: 0.845924 valid_1's rmse: 1.30434
[1200] training's rmse: 0.795976 valid_1's rmse: 1.30238
[1350] training's rmse: 0.749836 valid_1's rmse: 1.30117
[1500] training's rmse: 0.707757 valid_1's rmse: 1.30086
[1650] training's rmse: 0.668077 valid_1's rmse: 1.30066
Early stopping, best iteration is:
[1574] training's rmse: 0.688158 valid_1's rmse: 1.30025
Model training done in 16.82925057411194 seconds.
Fold error 1.3002515604803022
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39244 valid_1's rmse: 1.48006
[300] training's rmse: 1.20711 valid_1's rmse: 1.38671
[450] training's rmse: 1.09274 valid_1's rmse: 1.35321
[600] training's rmse: 1.00901 valid_1's rmse: 1.33819
[750] training's rmse: 0.940408 valid_1's rmse: 1.33118
[900] training's rmse: 0.879907 valid_1's rmse: 1.32724
[1050] training's rmse: 0.823912 valid_1's rmse: 1.32422
[1200] training's rmse: 0.77276 valid_1's rmse: 1.32251
[1350] training's rmse: 0.726346 valid_1's rmse: 1.32239
Early stopping, best iteration is:
[1280] training's rmse: 0.747705 valid_1's rmse: 1.32219
Model training done in 14.6046621799469 seconds.
Fold error 1.3221918261286292
Total error 1.3236443423346225
Total std 0.01119
Fitting big fold 16 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.3685 valid_1's rmse: 1.48504
[300] training's rmse: 1.16063 valid_1's rmse: 1.39258
[450] training's rmse: 1.0302 valid_1's rmse: 1.36039
[600] training's rmse: 0.9353 valid_1's rmse: 1.34887
[750] training's rmse: 0.856368 valid_1's rmse: 1.34248
[900] training's rmse: 0.789368 valid_1's rmse: 1.33996
[1050] training's rmse: 0.729959 valid_1's rmse: 1.33859
[1200] training's rmse: 0.673233 valid_1's rmse: 1.33747
[1350] training's rmse: 0.623557 valid_1's rmse: 1.33717
Early stopping, best iteration is:
[1396] training's rmse: 0.610025 valid_1's rmse: 1.33692
Model training done in 12.788934230804443 seconds.
Fold error 1.3369152717101305
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.4008 valid_1's rmse: 1.49499
[300] training's rmse: 1.22532 valid_1's rmse: 1.37511
[450] training's rmse: 1.11798 valid_1's rmse: 1.3285
[600] training's rmse: 1.04002 valid_1's rmse: 1.31127
[750] training's rmse: 0.975304 valid_1's rmse: 1.3029
[900] training's rmse: 0.917513 valid_1's rmse: 1.29942
[1050] training's rmse: 0.865771 valid_1's rmse: 1.29691
[1200] training's rmse: 0.818641 valid_1's rmse: 1.29522
[1350] training's rmse: 0.775639 valid_1's rmse: 1.29336
[1500] training's rmse: 0.735983 valid_1's rmse: 1.2929
[1650] training's rmse: 0.698317 valid_1's rmse: 1.2919
[1800] training's rmse: 0.6629 valid_1's rmse: 1.29147
Early stopping, best iteration is:
[1751] training's rmse: 0.674476 valid_1's rmse: 1.29106
Model training done in 18.51490569114685 seconds.
Fold error 1.2910597482563038
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.4018 valid_1's rmse: 1.46673
[300] training's rmse: 1.22405 valid_1's rmse: 1.36004
[450] training's rmse: 1.11532 valid_1's rmse: 1.3208
[600] training's rmse: 1.03658 valid_1's rmse: 1.3044
[750] training's rmse: 0.970812 valid_1's rmse: 1.29416
[900] training's rmse: 0.912856 valid_1's rmse: 1.28913
[1050] training's rmse: 0.860384 valid_1's rmse: 1.28499
[1200] training's rmse: 0.81233 valid_1's rmse: 1.28262
[1350] training's rmse: 0.767586 valid_1's rmse: 1.27979
[1500] training's rmse: 0.727263 valid_1's rmse: 1.27946
[1650] training's rmse: 0.689331 valid_1's rmse: 1.27848
[1800] training's rmse: 0.654348 valid_1's rmse: 1.27838
[1950] training's rmse: 0.620619 valid_1's rmse: 1.27794
Early stopping, best iteration is:
[1974] training's rmse: 0.615267 valid_1's rmse: 1.27783
Model training done in 24.019810914993286 seconds.
Fold error 1.2778268462864748
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39546 valid_1's rmse: 1.47874
[300] training's rmse: 1.21184 valid_1's rmse: 1.39009
[450] training's rmse: 1.09905 valid_1's rmse: 1.36208
[600] training's rmse: 1.01821 valid_1's rmse: 1.35081
[750] training's rmse: 0.953112 valid_1's rmse: 1.3451
[900] training's rmse: 0.895831 valid_1's rmse: 1.3416
[1050] training's rmse: 0.843766 valid_1's rmse: 1.33884
[1200] training's rmse: 0.796168 valid_1's rmse: 1.33799
[1350] training's rmse: 0.751738 valid_1's rmse: 1.33711
[1500] training's rmse: 0.711177 valid_1's rmse: 1.3365
[1650] training's rmse: 0.67169 valid_1's rmse: 1.33592
Early stopping, best iteration is:
[1673] training's rmse: 0.666085 valid_1's rmse: 1.33568
Model training done in 21.335248947143555 seconds.
Fold error 1.3356810037464442
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.3911 valid_1's rmse: 1.4945
[300] training's rmse: 1.20595 valid_1's rmse: 1.39958
[450] training's rmse: 1.09177 valid_1's rmse: 1.36655
[600] training's rmse: 1.00887 valid_1's rmse: 1.35547
[750] training's rmse: 0.940124 valid_1's rmse: 1.34912
[900] training's rmse: 0.879185 valid_1's rmse: 1.34512
[1050] training's rmse: 0.824766 valid_1's rmse: 1.3432
[1200] training's rmse: 0.774935 valid_1's rmse: 1.34184
[1350] training's rmse: 0.731258 valid_1's rmse: 1.34182
Early stopping, best iteration is:
[1303] training's rmse: 0.744782 valid_1's rmse: 1.34158
Model training done in 16.0803439617157 seconds.
Fold error 1.3415817992975496
Total error 1.3257806405435473
Total std 0.02667
Fitting big fold 17 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.38425 valid_1's rmse: 1.46403
[300] training's rmse: 1.17802 valid_1's rmse: 1.36997
[450] training's rmse: 1.04804 valid_1's rmse: 1.33693
[600] training's rmse: 0.951121 valid_1's rmse: 1.32551
[750] training's rmse: 0.870952 valid_1's rmse: 1.32001
[900] training's rmse: 0.801057 valid_1's rmse: 1.31742
[1050] training's rmse: 0.738264 valid_1's rmse: 1.31634
[1200] training's rmse: 0.684558 valid_1's rmse: 1.31502
[1350] training's rmse: 0.635932 valid_1's rmse: 1.3148
Early stopping, best iteration is:
[1369] training's rmse: 0.629998 valid_1's rmse: 1.31467
Model training done in 15.266962051391602 seconds.
Fold error 1.3146708110598608
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39741 valid_1's rmse: 1.50746
[300] training's rmse: 1.2184 valid_1's rmse: 1.41684
[450] training's rmse: 1.10942 valid_1's rmse: 1.38406
[600] training's rmse: 1.03085 valid_1's rmse: 1.37349
[750] training's rmse: 0.965894 valid_1's rmse: 1.37029
[900] training's rmse: 0.908158 valid_1's rmse: 1.36874
[1050] training's rmse: 0.856827 valid_1's rmse: 1.3681
[1200] training's rmse: 0.808275 valid_1's rmse: 1.36716
[1350] training's rmse: 0.763519 valid_1's rmse: 1.36621
Early stopping, best iteration is:
[1375] training's rmse: 0.756675 valid_1's rmse: 1.36566
Model training done in 17.771020889282227 seconds.
Fold error 1.3656570416515168
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39288 valid_1's rmse: 1.53091
[300] training's rmse: 1.21384 valid_1's rmse: 1.43199
[450] training's rmse: 1.10453 valid_1's rmse: 1.39817
[600] training's rmse: 1.02472 valid_1's rmse: 1.38388
[750] training's rmse: 0.95865 valid_1's rmse: 1.37619
[900] training's rmse: 0.900169 valid_1's rmse: 1.37175
[1050] training's rmse: 0.848392 valid_1's rmse: 1.36979
[1200] training's rmse: 0.801072 valid_1's rmse: 1.36853
[1350] training's rmse: 0.757279 valid_1's rmse: 1.36789
[1500] training's rmse: 0.718695 valid_1's rmse: 1.3671
Early stopping, best iteration is:
[1485] training's rmse: 0.722494 valid_1's rmse: 1.36693
Model training done in 19.066924810409546 seconds.
Fold error 1.3669255517484211
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39319 valid_1's rmse: 1.50155
[300] training's rmse: 1.2125 valid_1's rmse: 1.3955
[450] training's rmse: 1.10187 valid_1's rmse: 1.35759
[600] training's rmse: 1.02134 valid_1's rmse: 1.34234
[750] training's rmse: 0.954868 valid_1's rmse: 1.33409
[900] training's rmse: 0.896909 valid_1's rmse: 1.33008
[1050] training's rmse: 0.844577 valid_1's rmse: 1.32646
[1200] training's rmse: 0.795459 valid_1's rmse: 1.32387
[1350] training's rmse: 0.751969 valid_1's rmse: 1.32316
Early stopping, best iteration is:
[1299] training's rmse: 0.766396 valid_1's rmse: 1.32269
Model training done in 15.655407428741455 seconds.
Fold error 1.3226907123954845
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39808 valid_1's rmse: 1.46416
[300] training's rmse: 1.2125 valid_1's rmse: 1.36541
[450] training's rmse: 1.09793 valid_1's rmse: 1.3304
[600] training's rmse: 1.01502 valid_1's rmse: 1.31506
[750] training's rmse: 0.945425 valid_1's rmse: 1.30768
[900] training's rmse: 0.884973 valid_1's rmse: 1.30433
[1050] training's rmse: 0.830349 valid_1's rmse: 1.30274
[1200] training's rmse: 0.780368 valid_1's rmse: 1.30169
[1350] training's rmse: 0.734208 valid_1's rmse: 1.30069
[1500] training's rmse: 0.69201 valid_1's rmse: 1.29971
[1650] training's rmse: 0.651763 valid_1's rmse: 1.29895
[1800] training's rmse: 0.614991 valid_1's rmse: 1.29821
Early stopping, best iteration is:
[1769] training's rmse: 0.622125 valid_1's rmse: 1.29809
Model training done in 20.682893991470337 seconds.
Fold error 1.298089814956046
Total error 1.3250908016916014
Total std 0.02784
Fitting big fold 18 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.37802 valid_1's rmse: 1.48298
[300] training's rmse: 1.17149 valid_1's rmse: 1.3853
[450] training's rmse: 1.03973 valid_1's rmse: 1.34865
[600] training's rmse: 0.942872 valid_1's rmse: 1.33483
[750] training's rmse: 0.86513 valid_1's rmse: 1.32728
[900] training's rmse: 0.796482 valid_1's rmse: 1.3229
[1050] training's rmse: 0.733833 valid_1's rmse: 1.3194
[1200] training's rmse: 0.676047 valid_1's rmse: 1.31826
Early stopping, best iteration is:
[1220] training's rmse: 0.669279 valid_1's rmse: 1.31806
Model training done in 13.627465963363647 seconds.
Fold error 1.3180561006057108
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40293 valid_1's rmse: 1.46147
[300] training's rmse: 1.22524 valid_1's rmse: 1.35872
[450] training's rmse: 1.11749 valid_1's rmse: 1.32226
[600] training's rmse: 1.04025 valid_1's rmse: 1.30941
[750] training's rmse: 0.975093 valid_1's rmse: 1.30316
[900] training's rmse: 0.918152 valid_1's rmse: 1.29965
[1050] training's rmse: 0.866519 valid_1's rmse: 1.29777
Early stopping, best iteration is:
[1089] training's rmse: 0.853905 valid_1's rmse: 1.29737
Model training done in 17.44634485244751 seconds.
Fold error 1.297368284569688
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39277 valid_1's rmse: 1.51118
[300] training's rmse: 1.21106 valid_1's rmse: 1.42688
[450] training's rmse: 1.10108 valid_1's rmse: 1.40139
[600] training's rmse: 1.02187 valid_1's rmse: 1.39265
[750] training's rmse: 0.956851 valid_1's rmse: 1.38977
Early stopping, best iteration is:
[772] training's rmse: 0.94839 valid_1's rmse: 1.38917
Model training done in 11.983043193817139 seconds.
Fold error 1.389170621129051
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39575 valid_1's rmse: 1.49003
[300] training's rmse: 1.21492 valid_1's rmse: 1.39002
[450] training's rmse: 1.10418 valid_1's rmse: 1.35264
[600] training's rmse: 1.02324 valid_1's rmse: 1.33695
[750] training's rmse: 0.956392 valid_1's rmse: 1.32801
[900] training's rmse: 0.899109 valid_1's rmse: 1.32352
[1050] training's rmse: 0.847142 valid_1's rmse: 1.32145
[1200] training's rmse: 0.798086 valid_1's rmse: 1.31922
[1350] training's rmse: 0.753602 valid_1's rmse: 1.31737
[1500] training's rmse: 0.712468 valid_1's rmse: 1.31709
Early stopping, best iteration is:
[1534] training's rmse: 0.703343 valid_1's rmse: 1.31663
Model training done in 19.1785147190094 seconds.
Fold error 1.3166259346330778
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39238 valid_1's rmse: 1.4849
[300] training's rmse: 1.20743 valid_1's rmse: 1.38794
[450] training's rmse: 1.09432 valid_1's rmse: 1.3539
[600] training's rmse: 1.01126 valid_1's rmse: 1.33976
[750] training's rmse: 0.942444 valid_1's rmse: 1.3323
[900] training's rmse: 0.881686 valid_1's rmse: 1.32946
[1050] training's rmse: 0.828559 valid_1's rmse: 1.32775
[1200] training's rmse: 0.777715 valid_1's rmse: 1.32686
Early stopping, best iteration is:
[1234] training's rmse: 0.766796 valid_1's rmse: 1.32647
Model training done in 15.69185733795166 seconds.
Fold error 1.326467811585065
Total error 1.3266100923222426
Total std 0.0313
Fitting big fold 19 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.37209 valid_1's rmse: 1.50003
[300] training's rmse: 1.1739 valid_1's rmse: 1.39571
[450] training's rmse: 1.04618 valid_1's rmse: 1.35619
[600] training's rmse: 0.950742 valid_1's rmse: 1.3394
[750] training's rmse: 0.875044 valid_1's rmse: 1.33042
[900] training's rmse: 0.810508 valid_1's rmse: 1.3261
[1050] training's rmse: 0.752532 valid_1's rmse: 1.32236
[1200] training's rmse: 0.700162 valid_1's rmse: 1.32056
[1350] training's rmse: 0.651074 valid_1's rmse: 1.319
[1500] training's rmse: 0.606788 valid_1's rmse: 1.31729
[1650] training's rmse: 0.563561 valid_1's rmse: 1.31645
[1800] training's rmse: 0.52345 valid_1's rmse: 1.31568
[1950] training's rmse: 0.485924 valid_1's rmse: 1.31495
[2100] training's rmse: 0.451541 valid_1's rmse: 1.31495
Early stopping, best iteration is:
[2144] training's rmse: 0.4417 valid_1's rmse: 1.31476
Model training done in 21.724275827407837 seconds.
Fold error 1.314762918951203
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40479 valid_1's rmse: 1.44576
[300] training's rmse: 1.22799 valid_1's rmse: 1.33323
[450] training's rmse: 1.11998 valid_1's rmse: 1.28993
[600] training's rmse: 1.04099 valid_1's rmse: 1.27291
[750] training's rmse: 0.974912 valid_1's rmse: 1.26472
[900] training's rmse: 0.917683 valid_1's rmse: 1.25959
[1050] training's rmse: 0.864502 valid_1's rmse: 1.25648
[1200] training's rmse: 0.815792 valid_1's rmse: 1.25318
[1350] training's rmse: 0.771266 valid_1's rmse: 1.25083
[1500] training's rmse: 0.731069 valid_1's rmse: 1.2501
[1650] training's rmse: 0.692476 valid_1's rmse: 1.24904
Early stopping, best iteration is:
[1687] training's rmse: 0.683088 valid_1's rmse: 1.2488
Model training done in 21.510008096694946 seconds.
Fold error 1.2488003287965028
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39473 valid_1's rmse: 1.50151
[300] training's rmse: 1.21417 valid_1's rmse: 1.40825
[450] training's rmse: 1.10408 valid_1's rmse: 1.37682
[600] training's rmse: 1.02396 valid_1's rmse: 1.36585
[750] training's rmse: 0.958029 valid_1's rmse: 1.36244
[900] training's rmse: 0.89912 valid_1's rmse: 1.36084
[1050] training's rmse: 0.8476 valid_1's rmse: 1.35979
[1200] training's rmse: 0.79873 valid_1's rmse: 1.3582
[1350] training's rmse: 0.755195 valid_1's rmse: 1.3568
[1500] training's rmse: 0.715307 valid_1's rmse: 1.35584
[1650] training's rmse: 0.676833 valid_1's rmse: 1.35542
[1800] training's rmse: 0.640616 valid_1's rmse: 1.35506
[1950] training's rmse: 0.606335 valid_1's rmse: 1.35493
[2100] training's rmse: 0.574927 valid_1's rmse: 1.3549
Early stopping, best iteration is:
[2083] training's rmse: 0.578612 valid_1's rmse: 1.35461
Model training done in 25.629347562789917 seconds.
Fold error 1.3546124346744637
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.3957 valid_1's rmse: 1.47012
[300] training's rmse: 1.21065 valid_1's rmse: 1.39226
[450] training's rmse: 1.09875 valid_1's rmse: 1.36749
[600] training's rmse: 1.01742 valid_1's rmse: 1.35857
[750] training's rmse: 0.950718 valid_1's rmse: 1.35429
[900] training's rmse: 0.892272 valid_1's rmse: 1.35215
[1050] training's rmse: 0.840647 valid_1's rmse: 1.35051
[1200] training's rmse: 0.791766 valid_1's rmse: 1.34958
Early stopping, best iteration is:
[1163] training's rmse: 0.803168 valid_1's rmse: 1.34922
Model training done in 15.064796924591064 seconds.
Fold error 1.3492165451015403
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39213 valid_1's rmse: 1.48084
[300] training's rmse: 1.20565 valid_1's rmse: 1.39097
[450] training's rmse: 1.09148 valid_1's rmse: 1.35955
[600] training's rmse: 1.00938 valid_1's rmse: 1.34767
[750] training's rmse: 0.940706 valid_1's rmse: 1.34272
[900] training's rmse: 0.88055 valid_1's rmse: 1.33992
[1050] training's rmse: 0.82653 valid_1's rmse: 1.3396
[1200] training's rmse: 0.775266 valid_1's rmse: 1.33876
Early stopping, best iteration is:
[1246] training's rmse: 0.761031 valid_1's rmse: 1.33864
Model training done in 15.471073389053345 seconds.
Fold error 1.338638832820739
Total error 1.323754680376221
Total std 0.0387
Fitting big fold 20 out of 20
Fitting sub fold 1 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.36624 valid_1's rmse: 1.48729
[300] training's rmse: 1.15882 valid_1's rmse: 1.39517
[450] training's rmse: 1.02593 valid_1's rmse: 1.36271
[600] training's rmse: 0.928982 valid_1's rmse: 1.3511
[750] training's rmse: 0.847395 valid_1's rmse: 1.34496
[900] training's rmse: 0.777224 valid_1's rmse: 1.34182
[1050] training's rmse: 0.716115 valid_1's rmse: 1.34034
[1200] training's rmse: 0.661256 valid_1's rmse: 1.3392
[1350] training's rmse: 0.612082 valid_1's rmse: 1.33884
[1500] training's rmse: 0.566307 valid_1's rmse: 1.33823
[1650] training's rmse: 0.524941 valid_1's rmse: 1.33772
[1800] training's rmse: 0.488269 valid_1's rmse: 1.33751
Early stopping, best iteration is:
[1761] training's rmse: 0.497997 valid_1's rmse: 1.33732
Model training done in 18.59180212020874 seconds.
Fold error 1.3373157580926895
Fitting sub fold 2 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.3993 valid_1's rmse: 1.48479
[300] training's rmse: 1.22214 valid_1's rmse: 1.39169
[450] training's rmse: 1.11479 valid_1's rmse: 1.36017
[600] training's rmse: 1.03704 valid_1's rmse: 1.34611
[750] training's rmse: 0.971786 valid_1's rmse: 1.33727
[900] training's rmse: 0.914061 valid_1's rmse: 1.33438
[1050] training's rmse: 0.862704 valid_1's rmse: 1.33176
[1200] training's rmse: 0.8161 valid_1's rmse: 1.32984
[1350] training's rmse: 0.772611 valid_1's rmse: 1.32891
Early stopping, best iteration is:
[1372] training's rmse: 0.766802 valid_1's rmse: 1.32859
Model training done in 17.424221754074097 seconds.
Fold error 1.3285876218772839
Fitting sub fold 3 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.40222 valid_1's rmse: 1.46356
[300] training's rmse: 1.22367 valid_1's rmse: 1.35551
[450] training's rmse: 1.11499 valid_1's rmse: 1.31415
[600] training's rmse: 1.03611 valid_1's rmse: 1.29881
[750] training's rmse: 0.970221 valid_1's rmse: 1.28988
[900] training's rmse: 0.912253 valid_1's rmse: 1.28618
[1050] training's rmse: 0.860333 valid_1's rmse: 1.28293
[1200] training's rmse: 0.812912 valid_1's rmse: 1.28044
[1350] training's rmse: 0.769907 valid_1's rmse: 1.27968
[1500] training's rmse: 0.729599 valid_1's rmse: 1.2785
[1650] training's rmse: 0.690666 valid_1's rmse: 1.27786
Early stopping, best iteration is:
[1584] training's rmse: 0.707885 valid_1's rmse: 1.27763
Model training done in 19.63164210319519 seconds.
Fold error 1.2776275078106472
Fitting sub fold 4 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39429 valid_1's rmse: 1.50087
[300] training's rmse: 1.21255 valid_1's rmse: 1.39889
[450] training's rmse: 1.10151 valid_1's rmse: 1.3607
[600] training's rmse: 1.01956 valid_1's rmse: 1.3465
[750] training's rmse: 0.951831 valid_1's rmse: 1.33819
[900] training's rmse: 0.892624 valid_1's rmse: 1.33507
[1050] training's rmse: 0.838002 valid_1's rmse: 1.33323
[1200] training's rmse: 0.787748 valid_1's rmse: 1.33077
[1350] training's rmse: 0.742686 valid_1's rmse: 1.32965
[1500] training's rmse: 0.701872 valid_1's rmse: 1.32853
Early stopping, best iteration is:
[1477] training's rmse: 0.708296 valid_1's rmse: 1.32819
Model training done in 18.655680894851685 seconds.
Fold error 1.328185234095416
Fitting sub fold 5 out of 5
Training until validation scores don't improve for 100 rounds.
[150] training's rmse: 1.39565 valid_1's rmse: 1.47099
[300] training's rmse: 1.21226 valid_1's rmse: 1.37243
[450] training's rmse: 1.09844 valid_1's rmse: 1.33987
[600] training's rmse: 1.01566 valid_1's rmse: 1.32719
[750] training's rmse: 0.947424 valid_1's rmse: 1.32175
[900] training's rmse: 0.887997 valid_1's rmse: 1.31866
[1050] training's rmse: 0.835629 valid_1's rmse: 1.31752
[1200] training's rmse: 0.785573 valid_1's rmse: 1.31704
[1350] training's rmse: 0.740253 valid_1's rmse: 1.31714
Early stopping, best iteration is:
[1291] training's rmse: 0.757834 valid_1's rmse: 1.31676
Model training done in 15.674333572387695 seconds.
Fold error 1.3167582001027431
Total error 1.3233851538900245
Total std 0.02107
Wall time: 31min 26s
In [ ]:
print('Total error',np.mean(([np.mean(x) for x in fold_errors])))
print('Total std ',np.mean(([np.std (x) for x in fold_errors])))
In [60]:
print('Length of test predictions:', len(pred_test_list_lgb))
avg_pred_test_list_lgb = np.mean(pred_test_list_lgb, axis=0)
print('Length of avg test predictions:', len(avg_pred_test_list_lgb))
Length of test predictions: 20
Length of avg test predictions: 49342
In [62]:
# 20x oof train preds
with open(os.path.join(DATA_PATH, 'izmaylov_20folds_train_cv1323_std0021.pkl'), 'wb') as f:
pickle.dump(y_oof_lgb, f)
#20x test preds
with open(os.path.join(DATA_PATH, 'izmaylov_20folds_test_cv1323_std0021.pkl'), 'wb') as f:
pickle.dump(pred_test_list_lgb, f)
Content source: Diyago/Machine-Learning-scripts
Similar notebooks: