def main():
parser = ArgumentParser()
parser.add_argument('sources', help='sources', type=existing_directory)
parser.add_argument('--model',
help='model name',
default='model' + datetime.now().isoformat('T'))
args = parser.parse_args()
k = 2
images, labels = load_images_and_labels(args.sources)
train_data = images[len(images)//k:]
train_labels = labels[len(images)//k:]
test_data = images[:len(images)//k]
test_labels = labels[:len(labels)//k]
model.fit(train_data,
train_labels,
batch_size=batch_size,
epochs=epochs,
verbose=1,
shuffle=True,
validation_data=(test_data, test_labels))
with open('bin/{}.json'.format(args.model), 'w+') as file:
file.write(model.to_json())
model.save_weights('bin/{}.h5'.format(args.model))
from tensorflow import keras
logdir = "./folder" + datetime.now().strftime("%Y%m%d-%H%M%S")
tensorboard_callback = keras.callbacks.TensorBoard(log_dir=logdir)
from get_image_generator import train_generator, validation_generator, test_generator
from model import model
from meta_parameters import EPOCHS, STEPS_PER_EPOCH, VALIDATION_STEPS, BATCH_SIZE
model.fit_generator(
train_generator,
steps_per_epoch=STEPS_PER_EPOCH,
epochs=EPOCHS,
validation_data=validation_generator,
validation_steps=VALIDATION_STEPS,
use_multiprocessing=True,
callbacks=[tensorboard_callback],
)
model.save_weights('my_model_weights.h5')
print("evaluate", model.metrics_names)
print(model.evaluate_generator(
test_generator,
use_multiprocessing=True,
))
model.count_params()
model.summary()
#TODO numbe of parameters
# https://stackoverflow.com/questions/35792278/how-to-find-number-of-parameters-of-a-keras-model
from keras.optimizers import Adam
from nmt_utils import *
from model import model
m = 10000
Tx = 30
Ty = 10
n_a = 32
n_s = 64
learning_rate = 0.005
batch_size = 100
dataset, human_vocab, machine_vocab, inv_vocab = load_dataset(m)
X, Y, Xoh, Yoh = preprocess_data(dataset, human_vocab, machine_vocab, Tx, Ty)
model = model(Tx, Ty, n_a, n_s, len(human_vocab), len(machine_vocab))
# model.summary()
opt = Adam(lr=learning_rate, beta_1=0.9, beta_2=0.999, decay=0.001)
model.compile(loss='categorical_crossentropy',
optimizer=opt,
metrics=['accuracy'])
s0 = np.zeros((m, n_s))
c0 = np.zeros((m, n_s))
outputs = list(Yoh.swapaxes(0, 1))
model.fit([Xoh, s0, c0], outputs, epochs=50, batch_size=batch_size)
model.save_weights('models/model_50.h5')
reduce_lr_callback = tf.keras.callbacks.ReduceLROnPlateau(
monitor='loss',
factor=FLAGS['lr_factor'],
patience=FLAGS['patience'],
verbose=0,
mode='auto',
min_delta=FLAGS['min_delta'],
cooldown=0,
min_lr=FLAGS['min_lr'])
early_stop_callback = tf.keras.callbacks.EarlyStopping(monitor='loss',
patience=10)
EPOCHS = FLAGS['epochs']
history = model.fit(dataset,
epochs=EPOCHS,
callbacks=[checkpoint_callback, reduce_lr_callback])
end = datetime.now()
END_TIME = str(end).replace(' ', '_')[:-7]
training_time = str(end - start)
print('Training took {} hour/min/sec'.format(training_time.split('.')[0]))
# Save final model weights for freezing and exporting later
save_model_path = os.path.join(basedir, 'saved_models',
'final_{}'.format(END_TIME))
model.save_weights(save_model_path)
def on_epoch_end(self, epoch, logs=None):
# 保存最优
if logs['loss'] <= self.lowest:
self.lowest = logs['loss']
model.save_weights(config.BEST_MODEL_PATH)
print(utils.generate_random_poetry(tokenizer, model))
epochs = 100
batch_size = 256
dataset_size = x_train.shape[0]
printer = NeatPrinter(epoch=epochs)
indices = gen_indices(batch_size, dataset_size)
for i, epoch in enumerate(range(epochs), 1):
loss = []
accuracy = []
x_train, y_train = shuffle(x_train, y_train)
for (a, b) in indices:
x_true = x_train[a:b]
y_true = y_train[a:b]
with tf.GradientTape() as tape:
y_pred = model(x_true, training=True)
loss_value = loss_fn(y_true, y_pred)
loss.append(loss_value.numpy())
acc = accuracy_fn(y_true, y_pred)
accuracy.append(acc.numpy())
grads = tape.gradient(loss_value, model.trainable_weights)
optimizer.apply_gradients(zip(grads, model.trainable_weights))
loss_val = sum(loss) / len(loss)
acc_val = sum(accuracy) / len(accuracy)
printer.print(i, loss_val, acc_val)
model.save_weights('../MODEL DATA/tf-model.h5')
from model import model
from properties import *
from keras.callbacks import BaseLogger, ModelCheckpoint
from data_generator import midi_input_generator, generate_song_array, save_array_to_midi
print "Training model"
#print generate_song_array(model)
#save_array_to_midi([generate_song_array(model)], 'Generated_zero.mid')
cp = ModelCheckpoint(model_check_point_file,
monitor='loss',
verbose=1,
save_best_only=True,
mode='max')
model.fit_generator(midi_input_generator(),
num_seq_in_epoch,
nb_epoch=num_epochs,
callbacks=[BaseLogger(), cp])
model.save_weights('net_dump.nw')
save_array_to_midi([generate_song_array(model)], 'Generated.mid')
preprocess_function=preprocess_input)
reduce_lr_01 = ReduceLROnPlateau(monitor='val_1st_acc',
factor=0.2,
patience=5,
min_lr=0,
mode='max')
reduce_lr_02 = ReduceLROnPlateau(monitor='val_2nd_acc',
factor=0.2,
patience=5,
min_lr=0.,
mode='max')
reduce_lr_03 = ReduceLROnPlateau(monitor='val_3rd_acc',
factor=0.2,
patience=5,
min_lr=0,
mode='max')
reduce_lr_04 = ReduceLROnPlateau(monitor='val_4th_acc',
factor=0.2,
patience=5,
min_lr=0.,
mode='max')
for i in range(30):
f = model.fit_generator(
train_gen,
steps_per_epoch=7210 / 32,
epochs=1,
validation_data=test_gen,
validation_steps=1000 / 32,
callbacks=[reduce_lr_01, reduce_lr_02, reduce_lr_03, reduce_lr_04])
model.save_weights('weights/classifier.h5')
def save(model):
model.save_weights('my_model.h5')
def on_epoch_end(self, epoch, logs=None):
if logs['loss'] <= self.lowest:
self.lowest = logs['loss']
model.save_weights(settings.BEST_MODEL_PATH)
for i in range(settings.SHOW_NUM):
print(utils.generate_random_poetry(tokenizer, model))
return INIT_LR * 0.9**epoch
model = model()
model.fit(
x_train,
y_train, # prepared data
batch_size=BATCH_SIZE,
epochs=EPOCHS,
callbacks=[
keras.callbacks.LearningRateScheduler(lr_scheduler),
LrHistory(),
TqdmProgressCallback()
],
validation_data=(x_test, y_test),
shuffle=True,
verbose=0,
initial_epoch=0)
print('SAVING THE MODEL')
model.save_weights("weights.h5")
print('MODEL SAVED')
# make test predictions
y_pred_test = model.predict_proba(x_test)
y_pred_test_classes = np.argmax(y_pred_test, axis=1)
y_pred_test_max_probas = np.max(y_pred_test, axis=1)
EPOCHS = 20
train_gen = data_gen(train_dir, BATCH_SIZE)
val_gen = data_gen(test_dir, BATCH_SIZE)
train_gen.build_data()
val_gen.build_data()
now = jdatetime.datetime.today()
model_folder_name = '%s-%s[%s-%s]__Model' % (now.month, now.day, now.hour,
now.minute)
os.mkdir('data\\models\\%s' % model_folder_name)
tensorboard_callback = k.callbacks.TensorBoard(log_dir='data\\models\\%s' %
(model_folder_name),
histogram_freq=1)
adam = k.optimizers.Adam()
model.compile(loss='mean_squared_error', optimizer=adam)
history = model.fit_generator(
generator=train_gen.next_batch(),
steps_per_epoch=int(train_gen.n / train_gen.batch_size),
epochs=EPOCHS,
validation_data=val_gen.next_batch(),
validation_steps=int(val_gen.n / val_gen.batch_size),
verbose=1,
callbacks=[tensorboard_callback])
model.save_weights('data\\models\\%s\\model.h5' % model_folder_name)
imgs = np.mean(imgs_color, axis=1)
imgs = imgsGray[:, None, :, :]
# Train a little bit
nb_epoch = 100
mini_epoch = 10
num_steps = int(nb_epoch / mini_epoch)
for step in tqdm(range(0, num_steps)):
h = model.fit([speedx, imgs], {'steer_out': steer},
batch_size=32,
nb_epoch=mini_epoch,
verbose=1,
validation_split=0.1,
shuffle=True)
model.save_weights('steer_comma_{0}_{1:4.5}.h5'.format(
step, h.history['val_loss'][-1]),
overwrite=True)
# Make predictions
start = 25000
stop = 27000
preds = model.predict([speedx[start:stop], imgs[start:stop]])
steer_preds = preds.reshape([-1])
# Animation!
def get_point(s, start=0, end=63, height=16):
X = int(s * (end - start))
if X < start:
X = start
if X > end:
