def delete_token(self):
"""
Delete the token associated to the model
:return:
"""
if self.full_name_i is not None:
token_path = EPath('temp', f'token_{self.full_name}')
if token_path.exists():
token_path.unlink()
def delete_token_midis_path(self):
"""
:return:
"""
if self._save_midis_path_i is not None:
path = EPath(
'temp',
f'token_{self.full_name}-generation({self._save_midis_path_i}).txt'
)
if path.exists():
path.unlink()
def main(args):
"""
:param args:
:return:
"""
if not args.midi:
shutil.rmtree(path='generated_midis', ignore_errors=True)
if not args.hp:
shutil.rmtree(path='hp_search', ignore_errors=True)
if not args.model:
shutil.rmtree(path='saved_models', ignore_errors=True)
if not args.tensorboard:
shutil.rmtree(path='tensorboard', ignore_errors=True)
if not args.temp:
shutil.rmtree(path='temp', ignore_errors=True)
if not args.data_temp:
data_temp = EPath(g.path.get_data_folder_path(args)) / 'temp'
shutil.rmtree(path=data_temp, ignore_errors=True)
if not args.zip:
zip_path = EPath('my_zip.zip')
if zip_path.exists():
zip_path.unlink()
print('Done cleaning')
class CheckPoint(KerasCallback):
"""
"""
def __init__(self, filepath):
"""
:param filepath:
"""
self.filepath = EPath(filepath)
super(CheckPoint, self).__init__()
self.best_acc = -np.Inf
self.best_loss = np.Inf
def on_epoch_end(self, epoch, logs=None):
"""
:param epoch:
:param logs:
:return:
"""
logs = logs or {}
if self.greater_result(logs):
self.best_acc = self.get_val_acc_mean(logs)
self.best_loss = self.get_val_loss(logs)
if self.filepath.exists():
self.filepath.unlink()
# self.model.save_weights(self.filepath.as_posix(), overwrite=True)
# tf.keras.experimental.export_saved_model(self.model, self.filepath.as_posix())
with open(self.filepath, 'wb') as dump_file:
pickle.dump(dict(weights=self.model.get_weights()), dump_file)
def greater_result(self, logs):
"""
:param logs:
:return:
"""
acc = self.get_val_acc_mean(logs)
return np.greater(
acc, self.best_acc) or (np.equal(acc, self.best_acc) and np.less(
self.get_val_loss(logs), self.best_loss))
@staticmethod
def get_val_acc_mean(logs):
"""
:param logs:
:return:
"""
i = 0
res = 0
while f'val_Output_{i}_acc' in logs:
res += logs.get(f'val_Output_{i}_acc')
i += 1
return res / i
@staticmethod
def get_val_loss(logs):
"""
:param logs:
:return:
"""
return logs.get('val_loss')
