def do_cv_pred(train, test, files):
print("------- do preds --------")
ensemble_col = [f[1] for f in files]
train_x = train[ensemble_col]
test_x = test[ensemble_col].values.reshape(-1)
train_y = train["target"]
submission = pd.DataFrame()
submission["card_id"] = test["card_id"]
submission["target"] = 0
outliers = (train["target"] < -30).astype(int).values
split_num = 5
skf = model_selection.StratifiedKFold(n_splits=split_num,
shuffle=True,
random_state=4590)
train_preds = []
for idx, (train_index,
test_index) in enumerate(skf.split(train, outliers)):
X_train, X_test = train_x.iloc[train_index], train_x.iloc[
test_index]
y_train, y_test = train_y.iloc[train_index], train_y.iloc[
test_index]
reg = IsotonicRegression()
X_train = X_train.values.reshape(-1)
X_test = X_test.values.reshape(-1)
reg.fit(X_train, y_train)
valid_set_pred = reg.predict(X_test)
print(y_test.describe())
temp = pd.DataFrame(valid_set_pred)
print(temp.describe())
score = evaluator.rmse(y_test, valid_set_pred)
print(score)
y_pred = reg.predict(test_x)
submission["target"] = submission["target"] + y_pred
train_id = train.iloc[test_index]
train_cv_prediction = pd.DataFrame()
train_cv_prediction["card_id"] = train_id["card_id"]
train_cv_prediction["cv_pred"] = valid_set_pred
train_preds.append(train_cv_prediction)
train_output = pd.concat(train_preds, axis=0)
submission["target"] = submission["target"] / split_num
submission.to_csv(path_const.OUTPUT_SUB, index=False)
train_output["cv_pred"] = np.clip(train_output["cv_pred"], -33.219281,
18.0)
train_output.to_csv(path_const.OUTPUT_OOF, index=False)
df_pred = pd.merge(train[["card_id", "target"]],
train_output,
on="card_id")
rmse_score = evaluator.rmse(df_pred["target"], df_pred["cv_pred"])
print(rmse_score)
def do_cv_pred(train, test, files, use_cols=10, verbose=False):
print("------- do preds --------")
ensemble_col = [
f[1] for i, f in enumerate(files) if (i % 20) <= use_cols
]
if use_cols == 2:
print(ensemble_col)
train_x = train[ensemble_col]
test_x = test[ensemble_col]
train_y = train["target"]
submission = pd.DataFrame()
submission["card_id"] = test["card_id"]
submission["target"] = 0
outliers = (train["target"] < -30).astype(int).values
split_num = 5
skf = model_selection.StratifiedKFold(n_splits=split_num,
shuffle=True,
random_state=4590)
train_preds = []
for idx, (train_index,
test_index) in enumerate(skf.split(train, outliers)):
X_train, X_test = train_x.iloc[train_index], train_x.iloc[
test_index]
y_train, y_test = train_y.iloc[train_index], train_y.iloc[
test_index]
reg = BayesianRidge().fit(X_train, y_train)
valid_set_pred = reg.predict(X_test)
score = evaluator.rmse(y_test, valid_set_pred)
if verbose:
print(reg.coef_)
print(score)
y_pred = reg.predict(test_x)
submission["target"] = submission["target"] + y_pred
train_id = train.iloc[test_index]
train_cv_prediction = pd.DataFrame()
train_cv_prediction["card_id"] = train_id["card_id"]
train_cv_prediction["cv_pred"] = valid_set_pred
train_preds.append(train_cv_prediction)
train_output = pd.concat(train_preds, axis=0)
submission["target"] = submission["target"] / split_num
submission.to_csv(path_const.OUTPUT_SUB, index=False)
train_output["cv_pred"] = np.clip(train_output["cv_pred"], -33.219281,
18.0)
train_output.to_csv(path_const.OUTPUT_OOF, index=False)
df_pred = pd.merge(train[["card_id", "target"]],
train_output,
on="card_id")
rmse_score = evaluator.rmse(df_pred["target"], df_pred["cv_pred"])
print(rmse_score)
def do_cv_pred(train, test, files):
print("------- do preds --------")
ensemble_col = [f[1] for f in files] + ["has_new"]
train_x = train[ensemble_col]
test_x = test[ensemble_col]
train_y = train["target"]
submission = pd.DataFrame()
submission["card_id"] = test["card_id"]
submission["target"] = 0
outliers = (train["target"] < -30).astype(int).values
split_num = 5
skf = model_selection.StratifiedKFold(n_splits=split_num,
shuffle=True,
random_state=4590)
train_preds = []
lgb = pocket_lgb.ShallowLgb()
for idx, (train_index,
test_index) in enumerate(skf.split(train, outliers)):
X_train, X_test = train_x.iloc[train_index], train_x.iloc[
test_index]
y_train, y_test = train_y.iloc[train_index], train_y.iloc[
test_index]
model = lgb.do_train_direct(X_train, X_test, y_train, y_test)
valid_set_pred = model.predict(X_test)
score = evaluator.rmse(y_test, valid_set_pred)
print(score)
y_pred = model.predict(test_x)
submission["target"] = submission["target"] + y_pred
train_id = train.iloc[test_index]
train_cv_prediction = pd.DataFrame()
train_cv_prediction["card_id"] = train_id["card_id"]
train_cv_prediction["cv_pred"] = valid_set_pred
train_preds.append(train_cv_prediction)
if idx == 0:
lgb.show_feature_importance(model)
train_output = pd.concat(train_preds, axis=0)
submission["target"] = submission["target"] / split_num
submission.to_csv(path_const.OUTPUT_SUB, index=False)
train_output["cv_pred"] = np.clip(train_output["cv_pred"], -33.219281,
18.0)
train_output.to_csv(path_const.OUTPUT_OOF, index=False)
df_pred = pd.merge(train[["card_id", "target"]],
train_output,
on="card_id")
rmse_score = evaluator.rmse(df_pred["target"], df_pred["cv_pred"])
print(rmse_score)
def do_preds(train, test, files):
train = train.sort_values(by="bin", ascending=False)
test = test.sort_values(by="bin", ascending=False)
train["bin"] = train["bin"] * -33.219281
test["bin"] = test["bin"] * -33.219281
print(train.head())
print(train.tail())
print("------- do preds --------")
ensemble_col = [f[1] for f in files]
train_x = train[ensemble_col].values.reshape(-1)
reg = IsotonicRegression()
reg.fit(train_x, train["target"])
y_pred = reg.predict(train_x)
score = evaluator.rmse(train["target"], y_pred)
print(score)
test_x = test[ensemble_col].values.reshape(-1)
y_pred = reg.predict(test_x)
sub = pd.DataFrame()
sub["card_id"] = test["card_id"]
sub["target"] = y_pred
print(train["target"].describe())
# print(train["big"].describe())
print(sub["target"].describe())
sub.to_csv(path_const.OUTPUT_ENS, index=False)
def do_preds(train, files):
print("------- do preds --------")
ensemble_col = [f[1] for f in files]
train_x = train[ensemble_col]
reg = Ridge().fit(train_x, train["target"])
print(reg.coef_)
sig_idx = list()
for idx, coef in enumerate(reg.coef_):
if coef > 0.15:
sig_idx.append(idx)
print(sig_idx)
y_pred = reg.predict(train_x)
score = evaluator.rmse(train["target"], y_pred)
print(score)
return sig_idx
def do_preds(train, files):
print("------- do preds --------")
ensemble_col = [f[1] for f in files]
train_x = train[ensemble_col]
reg = Ridge().fit(train_x, train["target"])
print(reg.coef_)
sig_idx = list()
for idx, coef in enumerate(reg.coef_):
sig_idx.append((idx, coef))
sorted_idx = sorted(sig_idx, key=lambda tup: abs(tup[1]), reverse=True)
for i in sorted_idx:
print(i)
y_pred = reg.predict(train_x)
score = evaluator.rmse(train["target"], y_pred)
print(score)
return sorted_idx
def do_preds(train, test, files):
print("------- do preds --------")
ensemble_col = [f[1] for f in files]
train_x = train[ensemble_col]
reg = BayesianRidge().fit(train_x, train["target"])
print(reg.coef_)
y_pred = reg.predict(train_x)
score = evaluator.rmse(train["target"], y_pred)
print(score)
test_x = test[ensemble_col]
y_pred = reg.predict(test_x)
sub = pd.DataFrame()
sub["card_id"] = test["card_id"]
sub["target"] = y_pred
print(train["target"].describe())
# print(train["big"].describe())
print(sub["target"].describe())
sub.to_csv(path_const.OUTPUT_ENS, index=False)
def get_cv_score(param):
local_timer = pocket_timer.GoldenTimer(logger)
lgb = pocket_lgb.OptLgb(param)
for train_index, test_index in skf.split(train, outliers):
X_train, X_test = train_x.iloc[train_index], train_x.iloc[test_index]
y_train, y_test = train_y.iloc[train_index], train_y.iloc[test_index]
model = lgb.do_train_direct(X_train, X_test, y_train, y_test)
valid_set_pred = model.predict(X_test)
train_id = train.iloc[test_index]
train_cv_prediction = pd.DataFrame()
train_cv_prediction["card_id"] = train_id["card_id"]
train_cv_prediction["cv_pred"] = valid_set_pred
train_preds.append(train_cv_prediction)
train_output = pd.concat(train_preds, axis=0)
local_timer.time("end train in ")
train_output = pd.merge(train_output, train, on="card_id", how="left")
score = evaluator.rmse(train_output["target"], train_output["cv_pred"])
return score
def do_preds(train, test, files):
print("------- do preds --------")
ensemble_col = [f[1] for f in files]
train_x = train[ensemble_col]
reg = Ridge().fit(train_x, train["target"])
print(reg.coef_)
sig_idx = list()
for idx, coef in enumerate(reg.coef_):
if coef > 0.15:
sig_idx.append(idx)
print(sig_idx)
y_pred = reg.predict(train_x)
score = evaluator.rmse(train["target"], y_pred)
print(score)
test_x = test[ensemble_col]
y_pred = reg.predict(test_x)
sub = pd.DataFrame()
sub["card_id"] = test["card_id"]
sub["target"] = y_pred
print(train["target"].describe())
# print(train["big"].describe())
print(sub["target"].describe())
sub.to_csv(path_const.OUTPUT_ENS, index=False)
def print_score(train, files):
for f in files:
score = evaluator.rmse(train["target"], train[f[1]])
print(score)
timer = pocket_timer.GoldenTimer(logger)
csv_io = pocket_file_io.GoldenCsv()
data = input_loader.GoldenLoader().load_small_pred_new()
train, test, train_x, train_y, test_x = data
print(train_x.shape)
print(train_y.shape)
print(test_x.shape)
print(train_y.describe())
train_y = train_y.fillna(60)
print(train_x.columns)
timer.time("load csv")
mean_val = 6.040527
train_x["mean_val"] = mean_val
rmse_score = evaluator.rmse(train_y, train_x["mean_val"])
print(rmse_score)
pred_col_name = "pred_new"
submission = pd.DataFrame()
submission["card_id"] = test["card_id"]
submission[pred_col_name] = 0
train_cv = pd.DataFrame()
train_cv["card_id"] = train["card_id"]
train_cv[pred_col_name] = 0
outliers = (train["target"] < -30).astype(int).values
bagging_num = 1
split_num = 5
random_state = 4590
for bagging_index in range(bagging_num):
def merge_it(org, df):
ret_df = pd.merge(org, df, on="card_id", how="left")
return ret_df
new = csv_io.read_file(path_const.FEAT_FROM_TS_NEW)
old = csv_io.read_file(path_const.FEAT_FROM_TS_OLD)
new2 = csv_io.read_file(path_const.FEAT_FROM_TS_NEW2)
old2 = csv_io.read_file(path_const.FEAT_FROM_TS_OLD2)
train = csv_io.read_file(path_const.ORG_TRAIN)[["card_id", "target"]]
train = merge_it(train, new)
train = merge_it(train, old)
train = merge_it(train, new2)
train = merge_it(train, old2)
eval_cols = [
'pred_from_new_ts_mean',
'pred_from_old_ts_mean',
'pred_from_new_ts2_mean',
'pred_from_old_ts2_mean',
]
print(train.describe())
print(train.shape)
train = train[train["pred_from_new_ts2_mean"].notnull()]
print(train.shape)
for c in eval_cols:
score = evaluator.rmse(train["target"], train[c])
print(c, "=", score)
sub_with_outlier.loc[index,
'target'] = sub_with_outlier.loc[index, 'target'] - 5
index = (sub_with_outlier['target2'] <
-10) & (sub_with_outlier['target2'] >
-15) & (sub_with_outlier['target'] != outlier_value)
sub_with_outlier.loc[index,
'target'] = sub_with_outlier.loc[index, 'target'] - 1
index = (logistic_test['target'] > 0.94) & (sub_with_outlier['target'] !=
outlier_value)
sub_with_outlier.loc[index, 'target'] = outlier_value
index = sub_with_outlier['target'] <= outlier_value
sub_with_outlier.loc[index, 'target'] = outlier_value
sub_with_outlier[['card_id', 'target']].to_csv("../post/first_pp.csv",
index=False)
print(sub_with_outlier.shape)
print(sub_with_outlier.describe())
_train = train[["card_id", "target"]].copy()
_train.columns = ["card_id", "y_true"]
sub_with_outlier = pd.merge(sub_with_outlier,
_train,
on="card_id",
how="inner")
print(sub_with_outlier.shape)
score = evaluator.rmse(sub_with_outlier["target"], sub_with_outlier["y_true"])
print(score)
def do_cv_pred(train, test, files):
train_has_new_idx = train[train["has_new"] == 1].index
train_no_new_idx = train[train["has_new"] == 0].index
test_has_new_idx = test[test["has_new"] == 1].index
test_no_new_idx = test[test["has_new"] == 0].index
print("------- do preds --------")
ensemble_col = [f[1] for f in files]
train_x = train[ensemble_col]
test_x = test[ensemble_col]
has_new_test_x = test_x.iloc[test_has_new_idx]
no_new_test_x = test_x.iloc[test_no_new_idx]
train_y = train["target"]
has_new_sub = pd.DataFrame()
has_new_sub["card_id"] = test[test["has_new"] == 1]["card_id"]
has_new_sub["target"] = 0
no_new_sub = pd.DataFrame()
no_new_sub["card_id"] = test[test["has_new"] == 0]["card_id"]
no_new_sub["target"] = 0
outliers = (train["target"] < -30).astype(int).values
split_num = 5
skf = model_selection.StratifiedKFold(n_splits=split_num, shuffle=True, random_state=4590)
train_preds = []
for idx, (train_index, test_index) in enumerate(skf.split(train, outliers)):
# has new
_train_idx = [i for i in train_index if i in train_has_new_idx]
_test_idx = [i for i in test_index if i in train_has_new_idx]
X_train, X_test = train_x.iloc[_train_idx], train_x.iloc[_test_idx]
y_train, y_test = train_y.iloc[_train_idx], train_y.iloc[_test_idx]
reg = BayesianRidge().fit(X_train, y_train)
print(reg.coef_)
has_new_v_pred = reg.predict(X_test)
score = evaluator.rmse(y_test, has_new_v_pred)
print(score)
has_new_y_pred = reg.predict(has_new_test_x)
has_new_sub["target"] = has_new_sub["target"] + has_new_y_pred
train_id = train.iloc[_test_idx]
train_cv_prediction = pd.DataFrame()
train_cv_prediction["card_id"] = train_id["card_id"]
train_cv_prediction["cv_pred"] = has_new_v_pred
train_preds.append(train_cv_prediction)
# no new
_train_idx = [i for i in train_index if i in train_no_new_idx]
_test_idx = [i for i in test_index if i in train_no_new_idx]
X_train, X_test = train_x.iloc[_train_idx], train_x.iloc[_test_idx]
y_train, y_test = train_y.iloc[_train_idx], train_y.iloc[_test_idx]
reg = BayesianRidge().fit(X_train, y_train)
print(reg.coef_)
no_new_v_pred = reg.predict(X_test)
score = evaluator.rmse(y_test, no_new_v_pred)
print(score)
no_new_y_pred = reg.predict(no_new_test_x)
no_new_sub["target"] = no_new_sub["target"] + no_new_y_pred
train_id = train.iloc[_test_idx]
train_cv_prediction = pd.DataFrame()
train_cv_prediction["card_id"] = train_id["card_id"]
train_cv_prediction["cv_pred"] = no_new_v_pred
train_preds.append(train_cv_prediction)
train_output = pd.concat(train_preds, axis=0)
has_new_sub["target"] = has_new_sub["target"] / split_num
no_new_sub["target"] = no_new_sub["target"] / split_num
submission = pd.concat([has_new_sub, no_new_sub], axis=0)
submission.to_csv(path_const.OUTPUT_SUB, index=False)
train_output["cv_pred"] = np.clip(train_output["cv_pred"], -33.219281, 18.0)
train_output.to_csv(path_const.OUTPUT_OOF, index=False)
df_pred = pd.merge(train[["card_id", "target"]], train_output, on="card_id")
rmse_score = evaluator.rmse(df_pred["target"], df_pred["cv_pred"])
print(rmse_score)
train3.columns = ["card_id", "mlp"]
# train4.columns = ["card_id", "mlp4"]
train5.columns = ["card_id", "mlp_rank"]
train = pd.merge(train, train1, on="card_id", how="inner")
train = pd.merge(train, train2, on="card_id", how="inner")
train = pd.merge(train, train3, on="card_id", how="inner")
# train = pd.merge(train, train4, on="card_id", how="inner")
train = pd.merge(train, train5, on="card_id", how="inner")
print(train.shape)
print("-----")
print("co-eff...")
print(train[["target", "big", "small", "mlp", "mlp4", "mlp_rank"]].corr())
print("before score..")
score = evaluator.rmse(train["target"], train["big"])
print(score)
score = evaluator.rmse(train["target"], train["small"])
print(score)
score = evaluator.rmse(train["target"], train["mlp"])
print(score)
score = evaluator.rmse(train["target"], train["mlp4"])
print(score)
score = evaluator.rmse(train["target"], train["mlp_rank"])
print(score)
print("-----")
ensemble_col = ["big", "small", "mlp", "mlp4", "mlp_rank"]
train_x = train[ensemble_col]
reg = LinearRegression().fit(train_x, train["target"])
print(reg.coef_)
})
train_out_pred = pd.DataFrame({
"card_id": _outlier["card_id"],
"cv_pred": outlier_pred
})
train_cv_prediction = pd.concat([train_non_out_pred, train_out_pred], axis=0)
train_preds.append(train_cv_prediction)
timer.time("done one set in")
lgb.show_feature_importance(models[0], path_const.FEATURE_GAIN)
avg_score = str(total_score / split_num)
logger.print("average score= " + avg_score)
timer.time("end train in ")
submission["target"] = submission["target"] / split_num
submission.to_csv(path_const.OUTPUT_SUB, index=False)
train_output = pd.concat(train_preds, axis=0)
train_output.to_csv(path_const.OUTPUT_OOF, index=False)
y_true = train["target"]
y_pred = train_output["cv_pred"]
rmse_score = evaluator.rmse(y_true, y_pred)
logger.print("evaluator rmse score= " + str(rmse_score))
print(train["target"].describe())
logger.print(train_output.describe())
logger.print(submission.describe())
timer.time("done submission in ")
from elo.common import pocket_timer, pocket_logger, pocket_file_io, path_const
from elo.common import pocket_lgb, evaluator
from elo.utils import drop_col_util
from sklearn import model_selection
logger = pocket_logger.get_my_logger()
timer = pocket_timer.GoldenTimer(logger)
csv_io = pocket_file_io.GoldenCsv()
train = csv_io.read_file(path_const.ORG_TRAIN)[["card_id", "target"]]
train1 = csv_io.read_file("../sub/big_oof.csv")
test1 = csv_io.read_file("../sub/big_sub.csv")
train2 = csv_io.read_file("../sub/small_oof.csv")
test2 = csv_io.read_file("../sub/small_sub.csv")
timer.time("load csv in ")
train1.columns = ["card_id", "big"]
train2.columns = ["card_id", "small"]
train = pd.merge(train, train1, on="card_id", how="inner")
train = pd.merge(train, train2, on="card_id", how="inner")
train["avg"] = (train["big"] + train["small"]) / 2
score = evaluator.rmse(train["target"], train["big"])
print(score)
score = evaluator.rmse(train["target"], train["small"])
print(score)
score = evaluator.rmse(train["target"], train["avg"])
print(score)
def try_some(train, test):
train_x, test_x = train[pred_col], test[pred_col]
train_y = train["target"]
print(train_x.shape)
print(train_y.shape)
print(test_x.shape)
submission = pd.DataFrame()
submission["card_id"] = test["card_id"]
submission["target"] = 0
train_cv = pd.DataFrame()
train_cv["card_id"] = train["card_id"]
train_cv["cv_pred"] = 0
outliers = (train["target"] < -30).astype(int).values
bagging_num = 1
split_num = 5
random_state = 4590
for bagging_index in range(bagging_num):
skf = model_selection.StratifiedKFold(n_splits=split_num,
shuffle=True,
random_state=random_state)
logger.print("random_state=" + str(random_state))
lgb = pocket_lgb.GoldenLgb()
total_score = 0
models = []
train_preds = []
for train_index, test_index in skf.split(train, outliers):
X_train, X_test = train_x.iloc[train_index], train_x.iloc[
test_index]
y_train, y_test = train_y.iloc[train_index], train_y.iloc[
test_index]
model = lgb.do_train_direct(X_train, X_test, y_train, y_test)
score = model.best_score["valid_0"]["rmse"]
total_score += score
y_pred = model.predict(test_x)
valid_set_pred = model.predict(X_test)
models.append(model)
submission["target"] = submission["target"] + y_pred
train_id = train.iloc[test_index]
train_cv_prediction = pd.DataFrame()
train_cv_prediction["card_id"] = train_id["card_id"]
train_cv_prediction["cv_pred"] = valid_set_pred
train_preds.append(train_cv_prediction)
timer.time("done one set in")
train_output = pd.concat(train_preds, axis=0)
train_cv["cv_pred"] += train_output["cv_pred"]
lgb.show_feature_importance(models[0], path_const.FEATURE_GAIN)
avg_score = str(total_score / split_num)
logger.print("average score= " + avg_score)
timer.time("end train in ")
submission["target"] = submission["target"] / (bagging_num * split_num)
submission.to_csv(path_const.OUTPUT_SUB, index=False)
train_cv["cv_pred"] = train_cv["cv_pred"] / bagging_num
train_cv.to_csv(path_const.OUTPUT_OOF, index=False)
y_true = train_y
y_pred = train_cv["cv_pred"]
rmse_score = evaluator.rmse(y_true, y_pred)
logger.print("evaluator rmse score= " + str(rmse_score))
print(train["target"].describe())
logger.print(train_cv.describe())
logger.print(submission.describe())
timer.time("done submission in ")
def do_cv(self, data):
for d in data:
print(d.shape)
train, test, train_x, train_y, test_x = data
timer = pocket_timer.GoldenTimer(self.logger)
submission = pd.DataFrame()
submission["card_id"] = test["card_id"]
submission["target"] = 0
train_cv = pd.DataFrame()
train_cv["card_id"] = train["card_id"]
train_cv["cv_pred"] = 0
outliers = (train["target"] < -30).astype(int).values
bagging_num = 1
split_num = 5
for bagging_index in range(bagging_num):
skf = model_selection.StratifiedKFold(n_splits=split_num, shuffle=True, random_state=4590)
total_score = 0
train_preds = []
for idx, (train_index, test_index) in enumerate(skf.split(train, outliers)):
lr_schedule = learning_rate.GoldenLearningRate(0.1, 50).cosine_annealing_scheduler()
mlp = pocket_network2.GoldenMlp2(self.epochs, self.batch_size, lr_schedule)
network = mlp.build_model(train_x.shape[1])
X_train, X_test = train_x.iloc[train_index], train_x.iloc[test_index]
y_train, y_test = train_y.iloc[train_index], train_y.iloc[test_index]
print("start train")
model, history = mlp.do_train_direct(str(idx), network, X_train, X_test, y_train, y_test)
mlp.save_history(history, str(idx))
print('Loading Best Model')
model.load_weights(path_const.get_weight_file(str(idx)))
y_pred = model.predict(test_x, batch_size=self.batch_size)
y_pred = np.reshape(y_pred, -1)
y_pred = np.clip(y_pred, -33.219281, 18.0)
valid_set_pred = model.predict(X_test, batch_size=self.batch_size)
score = evaluator.rmse(y_test, valid_set_pred)
print(score)
total_score += score
submission["target"] = submission["target"] + y_pred
train_id = train.iloc[test_index]
train_cv_prediction = pd.DataFrame()
train_cv_prediction["card_id"] = train_id["card_id"]
train_cv_prediction["cv_pred"] = valid_set_pred
train_preds.append(train_cv_prediction)
timer.time("done one set in")
train_output = pd.concat(train_preds, axis=0)
train_cv["cv_pred"] += train_output["cv_pred"]
avg_score = str(total_score / split_num)
self.logger.print("average score= " + avg_score)
timer.time("end train in ")
submission["target"] = submission["target"] / (bagging_num * split_num)
# submission["target"] = np.clip(submission["target"], -33.219281, 18.0)
submission.to_csv(path_const.OUTPUT_SUB, index=False)
train_cv["cv_pred"] = train_cv["cv_pred"] / bagging_num
train_cv["cv_pred"] = np.clip(train_cv["cv_pred"], -33.219281, 18.0)
train_cv.to_csv(path_const.OUTPUT_OOF, index=False)
y_true = train_y
y_pred = train_cv["cv_pred"]
rmse_score = evaluator.rmse(y_true, y_pred)
self.logger.print("evaluator rmse score= " + str(rmse_score))
print(train["target"].describe())
self.logger.print(train_cv.describe())
self.logger.print(submission.describe())
timer.time("done submission in ")
