elif MODEL == 'LSTM':
class_nums = 2
lstm = LSTM(class_nums, embeddings_matrix, maxlen=maxlen)
_ = lstm(inputs=inputs)
adam = tf.keras.optimizers.Adam(lr=0.001,
beta_1=0.9,
beta_2=0.999,
epsilon=1e-08)
rmsprop = tf.keras.optimizers.RMSprop(lr=0.001, rho=0.9, epsilon=1e-06)
lstm.compile(loss='categorical_crossentropy', optimizer=adam)
if MODE == 'train':
lstm.fit(input_train,
tr_one_hot,
batch_size=32,
epochs=20,
verbose=1,
shuffle=True)
# Save model
lstm.save_weights(r'./saved_model/LSTM/lstm_20.HDF5')
print('model saved.')
else:
# Load model
lstm.load_weights(r'./saved_model/LSTM/lstm_20.HDF5')
print('model loaded.')
pred = lstm.predict(input_validation)
pred = np.argmax(pred, axis=1)
print('On validate data: ')
metrics(y_test, pred)
model = LSTM()
model.compile(optimizer="adam",
loss="categorical_crossentropy",
metrics=["accuracy"])
cp_path = "training_2/cp-{epoch:04d}.h5py"
cp_dir = os.path.dirname(cp_path)
# Create a callback that saves the model's weights
cp_callback = tf.keras.callbacks.ModelCheckpoint(filepath=cp_path,
verbose=1,
save_weights_only=True,
period=1)
log_dir = "./runs"
tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir=log_dir)
print("Initiating Training.")
# model.fit(train_ds, epochs = 10, validation_data = val_ds, steps_per_epoch = train_ct, validation_steps = val_ct, shuffle = True, callbacks = [cp_callback])
model.fit(train_ds,
epochs=10,
validation_data=val_ds,
steps_per_epoch=train_ct,
validation_steps=val_ct,
shuffle=True,
callbacks=[cp_callback, tensorboard_callback])
model.save_weights("./training_1/model.h5py")
print("Finished Training.")
# export PATH=/usr/local/cuda-10.0/bin:/usr/local/cuda-10.0/NsightCompute-1.0${PATH:+:${PATH}}
# export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
def main(args):
with open(args.config, 'r') as f:
config = json.load(f)
if config['optimizer'] == 'SGD':
optimizer = SGD(lr=config['learning_rate'],
decay=config['learning_rate'] / config['epochs'],
momentum=config['momentum'])
else:
raise Exception('Unsupported optimizer: {}.'.format(
config['optimizer']))
model_name = str.lower(config['model'])
if model_name == 'lstm':
model = LSTM(config['input_length'], 2)
elif model_name == 'conv1d':
model = Conv1D(config['input_length'], 2)
elif model_name == 'conv2d':
model = Conv2D(config['input_length'], 2)
else:
raise Exception('Unsupported model: {}.'.format(config['model']))
model.compile(loss='binary_crossentropy',
optimizer=optimizer,
metrics=['accuracy'])
wav_paths = glob('{}/**'.format(args.data_dir), recursive=True)
wav_paths = [x for x in wav_paths if '.wav' in x]
classes = sorted(os.listdir(args.data_dir))
le = LabelEncoder()
le.fit(classes)
labels = [get_class(x, args.data_dir) for x in wav_paths]
labels = le.transform(labels)
print('CLASSES: ', list(le.classes_))
print(le.transform(list(le.classes_)))
wav_train, wav_val, label_train, label_val = train_test_split(
wav_paths,
labels,
test_size=config['validation_split'],
random_state=0)
tg = DataGenerator(wav_train,
label_train,
config['input_length'],
len(set(label_train)),
batch_size=config['batch_size'])
vg = DataGenerator(wav_val,
label_val,
config['input_length'],
len(set(label_val)),
batch_size=config['batch_size'])
output_sub_dir = os.path.join(args.output_dir, model_name,
datetime.now().strftime('%Y%m%d_%H%M%S'))
os.makedirs(output_sub_dir)
callbacks = [
EarlyStopping(monitor='val_loss',
patience=config['patience'],
restore_best_weights=True,
verbose=1),
ModelCheckpoint(filepath=os.path.join(
output_sub_dir, 'model.{epoch:02d}-{val_loss:.4f}.h5'),
monitor='val_loss',
save_best_only=True,
verbose=1),
CSVLogger(os.path.join(output_sub_dir, 'epochs.csv'))
]
model.fit(tg,
validation_data=vg,
epochs=config['epochs'],
verbose=1,
callbacks=callbacks)
