def main(input_filepath, output_filepath, config_file):
"""Runs data loading and cleaning and pre-processing scripts and
saves data in ../processed."""
logger = logging.getLogger(__name__)
logger.info('Loading training data set, setting up pipeline, tuning,'
'training and evaluating final model.')
# Parse config file
# config = parse_config(config_file)
# Load training data
X_train = pd.read_csv(input_filepath + '/X_train.csv')
y_train = pd.read_csv(input_filepath + '/y_train.csv').values.ravel()
# Pre-processing and modeling pipeline
cat_features = X_train.select_dtypes(exclude='float64').columns
num_features = X_train.select_dtypes(include='float64').columns
pipe = Pipeline([('preprocessing',
preprocessing_pipeline(cat_features, num_features)),
('model',
TransformedTargetRegressor(regressor=SVR(),
func=np.log1p,
inverse_func=np.expm1))])
# Tune or select model
# kf = KFold(config['modeling']['num_folds'], shuffle=True,
# random_state=rng).get_n_splits(X_train.values)
model = Model(model=pipe)
# Train model
model.train(X_train, y_train)
# Save model
model.save(output_filepath + model.name + '.pkl')
fake_img = postprocess(fake_img) # [b, h, w, 3]
for i in range(fake_img.shape[0]):
sample_name = os.path.join(
args.path, 'eval',
val_dataset.load_name(index))
imsave(fake_img[i], sample_name)
index += 1
eval_progbar.add(fake_img.shape[0])
score_dict = get_inpainting_metrics(config.eval_path,
os.path.join(
args.path, 'eval'),
logger,
fid_test=config.fid_test)
if config.save_best and 'fid' in score_dict:
if 'fid' not in best_score or best_score[
'fid'] >= score_dict['fid']:
best_score = score_dict.copy()
best_score['iteration'] = iteration
model.save(prefix='best_fid')
if iteration % config.save_iters == 0:
model.save(prefix='last')
if iteration >= config.max_iters:
keep_training = False
break
logger.info('Best score: ' + str(best_score))