S5_LGB_main.py

import warnings
warnings.simplefilter('ignore')

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import gc,os,random
import time,datetime
from tqdm import tqdm
from sklearn.preprocessing import LabelEncoder

from utils import *
from model import *
root = args.root
seed = args.seed


df = pd.read_feather(f'{root}/all_feature.feather')

train_y =  pd.read_csv(f'{root}/train_labels.csv')
train = df[:train_y.shape[0]]
train['target'] = train_y['target']
test = df[train_y.shape[0]:].reset_index(drop=True)
del df

print(train.shape,test.shape)

lgb_config = {
    'lgb_params':{
                  'objective' : 'binary',
                  'metric' : 'binary_logloss',
                  'boosting': 'dart',
                  'max_depth' : -1,
                  'num_leaves' : 64,
                  'learning_rate' : 0.035,
                  'bagging_freq': 5,
                  'bagging_fraction' : 0.75,
                  'feature_fraction' : 0.05,
                  'min_data_in_leaf': 256,
                  'max_bin': 63,
                  'min_data_in_bin': 256,
                  # 'min_sum_heassian_in_leaf': 10,
                  'tree_learner': 'serial',
                  'boost_from_average': 'false',
                  'lambda_l1' : 0.1,
                  'lambda_l2' : 30,
                  'num_threads': 24,
                  'verbosity' : 1,
    },
    'feature_name':[],
    'rounds':4500,
    'early_stopping_rounds':100,
    'verbose_eval':50,
    'folds':5,
    'seed':seed
}

lgb_config = {
    'lgb_params':{
                  'objective' : 'binary',
                  'metric' : 'binary_logloss',
                  'boosting': 'dart',
                  'max_depth' : -1,
                  'num_leaves' : 64,
                  'learning_rate' : 0.035,
                  'bagging_freq': 5,
                  'bagging_fraction' : 0.75,
                  'feature_fraction' : 0.05,
                  'min_data_in_leaf': 256,
                  'max_bin': 63,
                  'min_data_in_bin': 256,
                  # 'min_sum_heassian_in_leaf': 10,
                  'tree_learner': 'serial',
                  'boost_from_average': 'false',
                  'lambda_l1' : 0.1,
                  'lambda_l2' : 30,
                  'num_threads': 24,
                  'verbosity' : 1,
    },
    'feature_name':[col for col in train.columns if col not in [id_name,label_name,'S_2'] and 'skew' not in col and 'kurt' not in col and 'sub_mean' not in col and 'div_mean' not in col],
    'rounds':4500,
    'early_stopping_rounds':100,
    'verbose_eval':50,
    'folds':5,
    'seed':seed
}
lgb_config['feature_name'] = [col for col in train.columns if col not in [id_name,label_name,'S_2'] and 'target' not in col]
Lgb_train_and_predict(train,test,lgb_config,aug=None,run_id='LGB_with_manual_feature')

lgb_config['feature_name'] = [col for col in train.columns if col not in [id_name,label_name,'S_2']]
Lgb_train_and_predict(train,test,lgb_config,aug=None,run_id='LGB_with_manual_feature_and_series_oof')