type=int,
dest='early_stopping_rounds',
required=False,
help='Number of round for early stopping')
parser.add_argument('--xgbparams',
type=str,
dest='xgb_params',
required=False,
help='Parameters of XGBoost constructor')
args = parser.parse_args()
print("#### Started %s ####" % os.path.basename(__file__))
head_train, df_independent_train, df_dependent_train = read_csv_dataset(
args.train_dataset_filename, 1)
eval_set = [(df_independent_train, df_dependent_train)]
if args.val_dataset_filename is not None:
head_val, df_independent_val, df_dependent_val = read_csv_dataset(
args.val_dataset_filename, 1)
eval_set.append((df_independent_val, df_dependent_val))
xgb_kwargs = prepare_kwargs_for_regressor(args)
model = xgb.XGBRegressor(**xgb_kwargs)
start_time = time.time()
model.fit(df_independent_train,
df_dependent_train,
eval_set=eval_set,
eval_metric=args.val_metrics,
type=str,
dest='dumpout_path',
required=False,
help='Dump directory (directory to store metric values)')
parser.add_argument('--xgbparams',
type=str,
dest='xgb_params',
required=False,
help='Parameters of XGBoost constructor')
args = parser.parse_args()
print("#### Started %s ####" % os.path.basename(__file__))
head_train, df_independent_train, df_dependent_train = read_csv_dataset(
args.train_dataset_filename, args.num_of_dependent_columns)
xgb_kwargs = prepare_kwargs_for_regressor(args)
model = sklmo.MultiOutputRegressor(xgb.XGBRegressor(**xgb_kwargs))
start_time = time.time()
model.fit(df_independent_train, df_dependent_train)
elapsed_time = time.time() - start_time
print("Training time:", time.strftime("%H:%M:%S",
time.gmtime(elapsed_time)))
jl.dump(model, args.model_file)
print("Generated one-variable function xgboost model '%s'" %
args.model_file)
print("#### Terminated %s ####" % os.path.basename(__file__))
parser.add_argument(
'--measures',
type=str,
dest='measures',
required=False,
nargs='+',
default=[],
help=
'List of built-in sklearn regression metrics to compare prediction with input dataset'
)
args = parser.parse_args()
print("#### Started %s ####" % os.path.basename(__file__))
head, df_independent, df_dependent = read_csv_dataset(
args.df_prediction, 1)
model = jl.load(args.model_file)
start_time = time.time()
prediction = model.predict(df_independent)
elapsed_time = time.time() - start_time
print("Predicting time:",
time.strftime("%H:%M:%S", time.gmtime(elapsed_time)))
compute_measures(df_dependent, prediction)
save_prediction(df_independent, prediction)
print("#### Terminated %s ####" % os.path.basename(__file__))
parser.add_argument(
'--measures',
type=str,
dest='measures',
required=False,
nargs='+',
default=[],
help=
'List of built-in sklearn regression measures to compare prediction with input dataset'
)
args = parser.parse_args()
print("#### Started %s ####" % os.path.basename(__file__))
df_prediction = read_csv_dataset(args.df_prediction)
columns = df_prediction.columns.tolist()
final_model = pcr.load_model(args.model_file)
start_time = time.time()
prediction = pcr.predict_model(final_model, data=df_prediction)
elapsed_time = time.time() - start_time
print("Predicting time:",
time.strftime("%H:%M:%S", time.gmtime(elapsed_time)))
compute_measures(prediction)
save_prediction(columns, prediction)
print("#### Terminated %s ####" % os.path.basename(__file__))
parser.add_argument(
'--measures',
type=str,
dest='measures',
required=False,
nargs='+',
default=[],
help=
'List of built-in sklearn regression metrics to compare prediction with input dataset'
)
args = parser.parse_args()
print("#### Started %s ####" % os.path.basename(__file__))
head, df_independent, df_dependent = read_csv_dataset(
args.df_prediction, args.num_of_dependent_columns)
model = jl.load(args.model_file)
start_time = time.time()
prediction = model.predict(df_independent)
elapsed_time = time.time() - start_time
print("Predicting time:",
time.strftime("%H:%M:%S", time.gmtime(elapsed_time)))
compute_measures(df_dependent, prediction)
save_prediction(df_independent, prediction)
print("#### Terminated %s ####" % os.path.basename(__file__))