lighGBM and Catboost

from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X,y,train_size = 0.90, random_state =42)
X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, train_size = 0.85, random_state =42)
categorical_features_indices = np.where(X.dtypes != np.float)[0]

#catboost
import catboost
from catboost import MetricVisualizer, Pool, CatBoostRegressor, cv


train_pool = Pool(data=X_train, label=y_train, cat_features=categorical_features_indices)

val_pool = Pool(data=X_val, label=y_val, cat_features=categorical_features_indices)

# No need to init here
test_pool = Pool(data=X_test, label=y_test, cat_features=categorical_features_indices)

#params
params = {
   'iterations':900,
   'loss_function': 'RMSE',
   'learning_rate': 0.0109, #1 0.102,
   'depth': 6,
   'l2_leaf_reg': 6,
   
   'border_count': 7,
   'thread_count': 7,
   
   'bagging_temperature': 2,
   'random_strength': 2.23,
   'colsample_bylevel': 0.85,
   
   'custom_metric': ['MAPE', 'R2'], #  'MAPE'] # R2, Quantile,
   'eval_metric': 'R2', # 'MAPE', # Overfitting detection if `use_best_model` is enabled
   'random_seed': 41,
   
   # 'max_ctr_complexity': 2,
   #'logging_level': 'Silent',
   'use_best_model':False # Takes
}


# specify the training parameters
reg_model = CatBoostRegressor(**params)


reg_model.fit(train_pool, eval_set=val_pool, plot=True, verbose=100)
importances = reg_model.get_feature_importance(prettified=True)
X_test['predict_qty']=reg_model.predict(X_test)

import pickle
with open ('reg_model21dec','wb') as f:
  pickle.dump(reg_model21dec,f)
  
with open ('reg_model21dec','rb') as f:
model =pickle.load(f)
 
 
 
#lightGBM
import lightgbm as lgb

#labelencoding

for col in train.columns:
  if train[col].dtype == 'object':
    le = LabelEncoder()
    le.fit(list(train[col].astype(str).values)+list(test[col].astype(str).values))
    train[col] = le.transform(list(train[col].astype(str).values))
    test[col] = le.transform(list(train[col].astype(str).values))

# lgb hyper-parameters
params = {'metric': 'rmse',
          'num_leaves': 255,
          'learning_rate': 0.005,
          'feature_fraction': 0.75,
          'bagging_fraction': 0.75,
          'bagging_freq': 5,
          'force_col_wise' : True,
          'random_state': 10}

cat_features = ['shop_id', 'city', 'item_category_id', 'category', 'month']

# lgb train and valid dataset
dtrain = lgb.Dataset(X_train, y_train)
dvalid = lgb.Dataset(X_valid, y_valid)
 
# Train LightGBM model
lgb_model = lgb.train(params=params,
                      train_set=dtrain,
                      num_boost_round=1500,
                      valid_sets=(dtrain, dvalid),
                      early_stopping_rounds=150,
                      categorical_feature=cat_features,
                      verbose_eval=100) 

preds = lgb_model.predict(X_test).clip(0,20)

submission['item_cnt_month'] = preds
submission.to_csv('submission.csv', index=False)