xgb_params['eta'] = 0.02
xgb_params['seed'] = SEED
xgb_params['silent'] = True # does help
xgb_params['verbose_eval'] = False
xgb_params['nrounds'] = 5000
xgb_params['early_stopping_rounds'] = 100
xgb_params['max_depth'] = 6
xgb_params['min_child_weight'] = 1
xgb_params['colsample_bytree'] = 0.724
xgb_params['subsample'] = 0.925
xgb_params['gamma'] = 0.512
xgb_params['alpha'] = 8.6
xgb_params['lambda'] = 1
full_data, ntrain, ntest = data_preparation()
xgb_clf = XgbWrapper(seed=SEED, params=xgb_params)
results = cross_validate(
full_data=full_data,
clf=xgb_clf,
seed=SEED,
ntrain=ntrain,
ntest=ntest,
features=FEATURES,
target=TARGET,
nfolds=4,
)
sub, v06, v33, oof_score = results
sub_to_csv(sub, v06, v33, oof_score[0], oof_score[1],
os.path.basename(sys.argv[0]))
xgb_params['seed'] = SEED
xgb_params['silent'] = True # does help
xgb_params['verbose_eval'] = False
xgb_params['nrounds'] = 2000
xgb_params['early_stopping_rounds'] = 100
xgb_params['max_depth'] = 5
xgb_params['min_child_weight'] = 1.91
xgb_params['colsample_bytree'] = 0.920
xgb_params['subsample'] = 0.856
xgb_params['gamma'] = 0.718
xgb_params['alpha'] = 1.83
xgb_params['lambda'] = 9.79
xgb_params['rate_drop'] = 0.262
full_data, ntrain, ntest = data_preparation()
xgb_clf = XgbWrapper(seed=SEED, params=xgb_params)
results = cross_validate(
full_data=full_data,
clf=xgb_clf,
seed=SEED,
ntrain=ntrain,
ntest=ntest,
features=FEATURES,
target=TARGET,
nfolds=4,
)
sub, v06, v33, oof_score = results
sub_to_csv(sub, v06, v33, oof_score[0], oof_score[1], "xgb_f2_cv4")
full_data, ntrain, ntest, FEATURES, CAT_FEATS = prepare_data()
# Get parameters
try:
exe_type = str(sys.argv[1])
opt_type = int(sys.argv[2])
except:
exe_type = 'test'
opt_type = 0
opt_path = '../tuning/' + "_".join(MODEL_NAME.split('_')[:1]) + ".csv"
xgb_turn_params = dict(pd.read_csv(opt_type).iloc[opt_type, :])
xgb_params = {**xgb_pick_params[exe_type], **xgb_turn_params}
# Define model and get results
xgb_clf = XgbWrapper(seed=SEED, params=xgb_params)
results = cross_validate(
full_data=full_data,
clf=xgb_clf,
seed=SEED,
ntrain=ntrain,
ntest=ntest,
features=FEATURES,
target=TARGET,
nfolds=4,
)
sub, v06, v33, oof_score = results
# Save submission to file
sub_to_csv(sub, v06, v33, oof_score[0], oof_score[1], MODEL_NAME)