def fit(self, train_x, train_y, validation_data_fit, train_loop_num,
**kwargs):
val_x, val_y = validation_data_fit
epochs = 10
patience = 2
callbacks = [
tf.keras.callbacks.EarlyStopping(monitor='val_loss',
patience=patience)
]
self._model.fit(
train_x,
ohe2cat(train_y),
epochs=epochs,
callbacks=callbacks,
validation_data=(val_x, ohe2cat(val_y)),
verbose=1, # Logs once per epoch.
batch_size=32,
shuffle=True)
def fit(self, train_x, train_y, validation_data_fit, round_num, **kwargs):
val_x, val_y = validation_data_fit
# if train_loop_num == 1:
# patience = 2
# epochs = 3
# elif train_loop_num == 2:
# patience = 3
# epochs = 10
# elif train_loop_num < 10:
# patience = 4
# epochs = 16
# elif train_loop_num < 15:
# patience = 4
# epochs = 24
# else:
# patience = 8
# epochs = 32
patience = 2
# epochs = self.epoch_cnt + 3
epochs = 10
callbacks = [
tf.keras.callbacks.EarlyStopping(monitor='val_loss',
patience=patience)
]
self._model.fit(
train_x,
ohe2cat(train_y),
epochs=epochs,
callbacks=callbacks,
validation_data=(val_x, ohe2cat(val_y)),
# validation_split=0.2,
verbose=1, # Logs once per epoch.
batch_size=32,
shuffle=True,
# initial_epoch=self.epoch_cnt,
# use_multiprocessing=True
)
self.epoch_cnt += 3
def fit(self, train_x, train_y, validation_data_fit, train_loop_num, **kwargs):
val_x, val_y = validation_data_fit
epochs = 5
patience = 2
batch_size = 32
# over_batch = len(train_x) % batch_size
# append_idx = np.random.choice(np.arange(len(train_x)), size=batch_size-over_batch, replace=False)
# train_x = np.concatenate([train_x, train_x[append_idx]], axis=0)
# train_y = np.concatenate([train_y, train_y[append_idx]], axis=0)
callbacks = [tf.keras.callbacks.EarlyStopping(monitor='val_loss', patience=patience)]
self._model.fit(
train_x,
ohe2cat(train_y),
epochs=epochs,
callbacks=callbacks,
validation_data=(val_x, ohe2cat(val_y)),
verbose=1, # Logs once per epoch.
batch_size=batch_size,
shuffle=True
)
def fit(self, train_x, train_y, validation_data_fit, train_loop_num,
**kwargs):
val_x, val_y = validation_data_fit
# if train_loop_num == 1:
# patience = 2
# epochs = 8
# elif train_loop_num == 2:
# patience = 3
# epochs = 10
# elif train_loop_num < 10:
# patience = 4
# epochs = 16
# elif train_loop_num < 15:
# patience = 4
# epochs = 24
# else:
# patience = 8
# epochs = 32
epochs = 3
patience = 2
callbacks = [
tf.keras.callbacks.EarlyStopping(monitor='val_loss',
patience=patience)
]
self._model.fit(
train_x,
ohe2cat(train_y),
epochs=epochs,
callbacks=callbacks,
validation_data=(val_x, ohe2cat(val_y)),
verbose=1, # Logs once per epoch.
batch_size=32,
shuffle=True)
def fit(self, train_loop_num=1, **kwargs):
# select model first, inorder to use preprocess data method
self._pre_select_model(train_loop_num)
log('fit {} for {} times'.format(self._model_name,
self._cur_model_run_loop))
self._cur_model_run_loop += 1
# get data
if self._round_num == 0:
train_x, train_y, val_x, val_y = self._data_manager.get_train_data(
train_loop_num=train_loop_num,
model_num=self._model_num,
round_num=self._round_num,
use_new_train=self._use_new_train,
use_mfcc=self._use_mfcc)
self._is_nedd_30s = self._data_manager.need_30s
if self._is_nedd_30s:
self._use_mel_round = 3
else:
self._use_mel_round = 2
else:
if self._round_num == self._use_mel_round:
self._use_mfcc = False
else:
self._use_mfcc = True
train_x, train_y, val_x, val_y = self._data_manager.get_train_data(
train_loop_num=train_loop_num,
model_num=self._model_num,
round_num=self._round_num,
use_new_train=self._use_new_train,
use_mfcc=self._use_mfcc)
self._val_set = (val_x, val_y)
self._input_shape = train_x.shape
log('train_x: {}; train_y: {};'.format(train_x.shape, train_y.shape) +
' val_x: {}; val_y: {};'.format(val_x.shape, val_y.shape))
# init model really
self._get_or_create_model()
self._classes = np.unique(ohe2cat(train_y))
self._model.fit(train_x, train_y, (val_x, val_y), self._round_num,
**kwargs)
def fit(self, x_train, y_train, *args, **kwargs):
self._model.fit(x_train, ohe2cat(y_train))
