# Params
lgb_params = {
'objective': 'binary',
'boosting_type': 'gbdt',
'metric': METRIC,
'num_threads': N_THREADS,
'verbose': VERBOSE,
'seed': SEED,
'n_estimators': N_ESTIMATORS,
'early_stopping_rounds': EARLY_STOPPING_ROUNDS
}
logger = utility.get_logger(LOGGER_NAME, MODEL_NUMBER, run_id, LOG_DIR)
utility.set_seed(SEED)
logger.info(f'Running for Model Number {MODEL_NUMBER}')
utility.update_tracking(run_id,
"model_number",
MODEL_NUMBER,
drop_incomplete_rows=True)
utility.update_tracking(run_id, "model_type", MODEL_TYPE)
utility.update_tracking(run_id, "is_test", IS_TEST)
utility.update_tracking(run_id, "n_estimators", N_ESTIMATORS)
utility.update_tracking(run_id, "early_stopping_rounds", EARLY_STOPPING_ROUNDS)
utility.update_tracking(run_id, "random_state", SEED)
utility.update_tracking(run_id, "n_threads", N_THREADS)
#utility.update_tracking(run_id, "learning_rate", LEARNING_RATE)
utility.update_tracking(run_id, "n_fold", N_FOLDS)
import feather
import sys
import pandas as pd
import numpy as np
sys.path.insert(0, "/home/jupyter/kaggle/energy/src")
import utility
utility.set_seed(42)
# Read weather_train and weather_test data
_, _, weather_train_df, weather_test_df, building_df = utility.read_data(utility.CREATED_DATA_DIR,
train=False, test=False,
weather_train=True, weather_test=True,
building=True)
print(f'Shape of weather_train_df : {weather_train_df.shape}')
print(f'Shape of weather_test_df : {weather_test_df.shape}')
# columns_name = ['air_temperature', 'cloud_coverage', 'dew_temperature',
# 'precip_depth_1_hr', 'sea_level_pressure', 'wind_direction',
# 'wind_speed']
# cloud_coverage, precip_depth_1_hr, sea_level_pressure, wind_direction
# are failing now for some reason. Hence, filling only for three weather
# attributes
columns_name = ['air_temperature', 'dew_temperature', 'wind_speed']
print('Null distribution before filling')
print(weather_train_df[columns_name].isna().sum())
import utility
import data
import model
import loss
from option import args
from checkpoint import Checkpoint
from trainer import Trainer
utility.set_seed(args.seed) # 设置随机种子,方便结果复现
checkpoint = Checkpoint(args)
if checkpoint.ok:
loader = data.Data(args)
model = model.Model(args, checkpoint)
loss = loss.Loss(args, checkpoint) if not args.test_only else None
t = Trainer(args, loader, model, loss, checkpoint)
while not t.terminate():
t.train()
t.test()
checkpoint.done()
import utility
import data
import model
import loss
from option import args
from checkpoint import Checkpoint
from trainer import Trainer
utility.set_seed(args.seed)
checkpoint = Checkpoint(args)
if checkpoint.ok:
loader = data.Data(args)
model = model.Model(args, checkpoint)
loss = loss.Loss(args, checkpoint) if not args.test_only else None
t = Trainer(args, loader, model, loss, checkpoint)
while not t.terminate():
t.train()
t.test()
checkpoint.done()