TensorFlow 2.0 keras 트레이닝 모델 사용의 실현
19435 단어 TensorFlow2keras훈련 모형
1. 일반적인 모델 구조, 훈련, 테스트 절차
#
inputs = keras.Input(shape=(784,), name='mnist_input')
h1 = layers.Dense(64, activation='relu')(inputs)
h1 = layers.Dense(64, activation='relu')(h1)
outputs = layers.Dense(10, activation='softmax')(h1)
model = keras.Model(inputs, outputs)
# keras.utils.plot_model(model, 'net001.png', show_shapes=True)
model.compile(optimizer=keras.optimizers.RMSprop(),
loss=keras.losses.SparseCategoricalCrossentropy(),
metrics=[keras.metrics.SparseCategoricalAccuracy()])
#
(x_train, y_train), (x_test, y_test) = keras.datasets.mnist.load_data()
x_train = x_train.reshape(60000, 784).astype('float32') /255
x_test = x_test.reshape(10000, 784).astype('float32') /255
x_val = x_train[-10000:]
y_val = y_train[-10000:]
x_train = x_train[:-10000]
y_train = y_train[:-10000]
#
history = model.fit(x_train, y_train, batch_size=64, epochs=3,
validation_data=(x_val, y_val))
print('history:')
print(history.history)
result = model.evaluate(x_test, y_test, batch_size=128)
print('evaluate:')
print(result)
pred = model.predict(x_test[:2])
print('predict:')
print(pred)
2. 사용자 정의 손실 및 지표
사용자 정의 지표는 Metric 클래스를 계승하고 함수를 다시 쓰기만 하면 된다
_init_(self), 초기화.
update_state(self, y_true, y_pred, sample_weight = None), 대상 y_ 사용true 및 모델 예측 y_pred로 상태 변수를 업데이트합니다.
result (self), 상태 변수를 사용하여 최종 결과를 계산합니다.
reset_states(self), 도량의 상태를 초기화합니다.
# , CatgoricalTruePositives ,
class CatgoricalTruePostives(keras.metrics.Metric):
def __init__(self, name='binary_true_postives', **kwargs):
super(CatgoricalTruePostives, self).__init__(name=name, **kwargs)
self.true_postives = self.add_weight(name='tp', initializer='zeros')
def update_state(self, y_true, y_pred, sample_weight=None):
y_pred = tf.argmax(y_pred)
y_true = tf.equal(tf.cast(y_pred, tf.int32), tf.cast(y_true, tf.int32))
y_true = tf.cast(y_true, tf.float32)
if sample_weight is not None:
sample_weight = tf.cast(sample_weight, tf.float32)
y_true = tf.multiply(sample_weight, y_true)
return self.true_postives.assign_add(tf.reduce_sum(y_true))
def result(self):
return tf.identity(self.true_postives)
def reset_states(self):
self.true_postives.assign(0.)
model.compile(optimizer=keras.optimizers.RMSprop(1e-3),
loss=keras.losses.SparseCategoricalCrossentropy(),
metrics=[CatgoricalTruePostives()])
model.fit(x_train, y_train,
batch_size=64, epochs=3)
# loss
class ActivityRegularizationLayer(layers.Layer):
def call(self, inputs):
self.add_loss(tf.reduce_sum(inputs) * 0.1)
return inputs
inputs = keras.Input(shape=(784,), name='mnist_input')
h1 = layers.Dense(64, activation='relu')(inputs)
h1 = ActivityRegularizationLayer()(h1)
h1 = layers.Dense(64, activation='relu')(h1)
outputs = layers.Dense(10, activation='softmax')(h1)
model = keras.Model(inputs, outputs)
# keras.utils.plot_model(model, 'net001.png', show_shapes=True)
model.compile(optimizer=keras.optimizers.RMSprop(),
loss=keras.losses.SparseCategoricalCrossentropy(),
metrics=[keras.metrics.SparseCategoricalAccuracy()])
model.fit(x_train, y_train, batch_size=32, epochs=1)
# metric
class MetricLoggingLayer(layers.Layer):
def call(self, inputs):
self.add_metric(keras.backend.std(inputs),
name='std_of_activation',
aggregation='mean')
return inputs
inputs = keras.Input(shape=(784,), name='mnist_input')
h1 = layers.Dense(64, activation='relu')(inputs)
h1 = MetricLoggingLayer()(h1)
h1 = layers.Dense(64, activation='relu')(h1)
outputs = layers.Dense(10, activation='softmax')(h1)
model = keras.Model(inputs, outputs)
# keras.utils.plot_model(model, 'net001.png', show_shapes=True)
model.compile(optimizer=keras.optimizers.RMSprop(),
loss=keras.losses.SparseCategoricalCrossentropy(),
metrics=[keras.metrics.SparseCategoricalAccuracy()])
model.fit(x_train, y_train, batch_size=32, epochs=1)
# model
# metric
class MetricLoggingLayer(layers.Layer):
def call(self, inputs):
self.add_metric(keras.backend.std(inputs),
name='std_of_activation',
aggregation='mean')
return inputs
inputs = keras.Input(shape=(784,), name='mnist_input')
h1 = layers.Dense(64, activation='relu')(inputs)
h2 = layers.Dense(64, activation='relu')(h1)
outputs = layers.Dense(10, activation='softmax')(h2)
model = keras.Model(inputs, outputs)
model.add_metric(keras.backend.std(inputs),
name='std_of_activation',
aggregation='mean')
model.add_loss(tf.reduce_sum(h1)*0.1)
# keras.utils.plot_model(model, 'net001.png', show_shapes=True)
model.compile(optimizer=keras.optimizers.RMSprop(),
loss=keras.losses.SparseCategoricalCrossentropy(),
metrics=[keras.metrics.SparseCategoricalAccuracy()])
model.fit(x_train, y_train, batch_size=32, epochs=1)
처리 사용 validation_u데이터 전송 테스트 데이터,validation_u 사용 가능split 구분 검증 데이터ps:validation_split는numpy 데이터로 훈련된 상황에서만 사용할 수 있습니다
model.fit(x_train, y_train, batch_size=32, epochs=1, validation_split=0.2)
3. tf를 사용한다.데이터 구조 데이터
def get_compiled_model():
inputs = keras.Input(shape=(784,), name='mnist_input')
h1 = layers.Dense(64, activation='relu')(inputs)
h2 = layers.Dense(64, activation='relu')(h1)
outputs = layers.Dense(10, activation='softmax')(h2)
model = keras.Model(inputs, outputs)
model.compile(optimizer=keras.optimizers.RMSprop(),
loss=keras.losses.SparseCategoricalCrossentropy(),
metrics=[keras.metrics.SparseCategoricalAccuracy()])
return model
model = get_compiled_model()
train_dataset = tf.data.Dataset.from_tensor_slices((x_train, y_train))
train_dataset = train_dataset.shuffle(buffer_size=1024).batch(64)
val_dataset = tf.data.Dataset.from_tensor_slices((x_val, y_val))
val_dataset = val_dataset.batch(64)
# model.fit(train_dataset, epochs=3)
# steps_per_epoch epoch
# validation_steps ,
model.fit(train_dataset, epochs=3, steps_per_epoch=100,
validation_data=val_dataset, validation_steps=3)
4. 샘플 가중치와 클래스 가중치
'샘플 가중치'그룹은 일괄 처리에서 각 샘플이 총 손실을 계산할 때 얼마나 많은 가중치를 가져야 하는지를 지정하는 숫자 그룹입니다.그것은 통상적으로 불균형적인 분류 문제에 쓰인다.사용 권한이 1과 0일 때, 이 수조는 손실 함수의 마스크로 사용할 수 있다. (일부 견본이 총 손실에 기여한 바를 완전히 버리고)
'클래스 권중'dict는 같은 개념의 더욱 구체적인 실례이다. 클래스 인덱스를 이 클래스에 속하는 샘플에 사용할 샘플 권중에 비추는 것이다.예를 들어 클래스'0'이 데이터의 클래스'1'보다 두 배 적으면class_weight = {0:1.,1:0.5}.
# 5
import numpy as np
#
model = get_compiled_model()
class_weight = {i:1.0 for i in range(10)}
class_weight[5] = 2.0
print(class_weight)
model.fit(x_train, y_train,
class_weight=class_weight,
batch_size=64,
epochs=4)
#
model = get_compiled_model()
sample_weight = np.ones(shape=(len(y_train),))
sample_weight[y_train == 5] = 2.0
model.fit(x_train, y_train,
sample_weight=sample_weight,
batch_size=64,
epochs=4)
# tf.data
model = get_compiled_model()
sample_weight = np.ones(shape=(len(y_train),))
sample_weight[y_train == 5] = 2.0
train_dataset = tf.data.Dataset.from_tensor_slices((x_train, y_train,
sample_weight))
train_dataset = train_dataset.shuffle(buffer_size=1024).batch(64)
val_dataset = tf.data.Dataset.from_tensor_slices((x_val, y_val))
val_dataset = val_dataset.batch(64)
model.fit(train_dataset, epochs=3, )
5. 다수입 다수출 모델
image_input = keras.Input(shape=(32, 32, 3), name='img_input')
timeseries_input = keras.Input(shape=(None, 10), name='ts_input')
x1 = layers.Conv2D(3, 3)(image_input)
x1 = layers.GlobalMaxPooling2D()(x1)
x2 = layers.Conv1D(3, 3)(timeseries_input)
x2 = layers.GlobalMaxPooling1D()(x2)
x = layers.concatenate([x1, x2])
score_output = layers.Dense(1, name='score_output')(x)
class_output = layers.Dense(5, activation='softmax', name='class_output')(x)
model = keras.Model(inputs=[image_input, timeseries_input],
outputs=[score_output, class_output])
keras.utils.plot_model(model, 'multi_input_output_model.png'
, show_shapes=True)
# loss metrics
model.compile(
optimizer=keras.optimizers.RMSprop(1e-3),
loss=[keras.losses.MeanSquaredError(),
keras.losses.CategoricalCrossentropy()])
# loss
model.compile(
optimizer=keras.optimizers.RMSprop(1e-3),
loss={'score_output': keras.losses.MeanSquaredError(),
'class_output': keras.losses.CategoricalCrossentropy()},
metrics={'score_output': [keras.metrics.MeanAbsolutePercentageError(),
keras.metrics.MeanAbsoluteError()],
'class_output': [keras.metrics.CategoricalAccuracy()]},
loss_weight={'score_output': 2., 'class_output': 1.})
# loss 0
model.compile(
optimizer=keras.optimizers.RMSprop(1e-3),
loss=[None, keras.losses.CategoricalCrossentropy()])
# Or dict loss version
model.compile(
optimizer=keras.optimizers.RMSprop(1e-3),
loss={'class_output': keras.losses.CategoricalCrossentropy()})
6. 콜백 사용
Keras의 리셋은 훈련 기간(epoch가 시작될 때,batch가 끝날 때,epoch가 끝날 때 등)에 서로 다른 점에서 호출되는 객체로 다음과 같은 행동을 수행할 수 있습니다.
6.1 콜백 사용
model = get_compiled_model()
callbacks = [
keras.callbacks.EarlyStopping(
# Stop training when `val_loss` is no longer improving
monitor='val_loss',
# "no longer improving" being defined as "no better than 1e-2 less"
min_delta=1e-2,
# "no longer improving" being further defined as "for at least 2 epochs"
patience=2,
verbose=1)
]
model.fit(x_train, y_train,
epochs=20,
batch_size=64,
callbacks=callbacks,
validation_split=0.2)
# checkpoint
model = get_compiled_model()
check_callback = keras.callbacks.ModelCheckpoint(
filepath='mymodel_{epoch}.h5',
save_best_only=True,
monitor='val_loss',
verbose=1
)
model.fit(x_train, y_train,
epochs=3,
batch_size=64,
callbacks=[check_callback],
validation_split=0.2)
#
initial_learning_rate = 0.1
lr_schedule = keras.optimizers.schedules.ExponentialDecay(
initial_learning_rate,
decay_steps=10000,
decay_rate=0.96,
staircase=True
)
optimizer = keras.optimizers.RMSprop(learning_rate=lr_schedule)
# tensorboard
tensorboard_cbk = keras.callbacks.TensorBoard(log_dir='./full_path_to_your_logs')
model.fit(x_train, y_train,
epochs=5,
batch_size=64,
callbacks=[tensorboard_cbk],
validation_split=0.2)
6.2 자체 콜백 작성 방법
class LossHistory(keras.callbacks.Callback):
def on_train_begin(self, logs):
self.losses = []
def on_epoch_end(self, batch, logs):
self.losses.append(logs.get('loss'))
print('
loss:',self.losses[-1])
model = get_compiled_model()
callbacks = [
LossHistory()
]
model.fit(x_train, y_train,
epochs=3,
batch_size=64,
callbacks=callbacks,
validation_split=0.2)
7. 자체 구조 훈련과 검증 순환
# Get the model.
inputs = keras.Input(shape=(784,), name='digits')
x = layers.Dense(64, activation='relu', name='dense_1')(inputs)
x = layers.Dense(64, activation='relu', name='dense_2')(x)
outputs = layers.Dense(10, activation='softmax', name='predictions')(x)
model = keras.Model(inputs=inputs, outputs=outputs)
# Instantiate an optimizer.
optimizer = keras.optimizers.SGD(learning_rate=1e-3)
# Instantiate a loss function.
loss_fn = keras.losses.SparseCategoricalCrossentropy()
# Prepare the training dataset.
batch_size = 64
train_dataset = tf.data.Dataset.from_tensor_slices((x_train, y_train))
train_dataset = train_dataset.shuffle(buffer_size=1024).batch(batch_size)
#
for epoch in range(3):
print('epoch: ', epoch)
for step, (x_batch_train, y_batch_train) in enumerate(train_dataset):
# gradient tape,
with tf.GradientTape() as tape:
logits = model(x_batch_train)
loss_value = loss_fn(y_batch_train, logits)
grads = tape.gradient(loss_value, model.trainable_variables)
optimizer.apply_gradients(zip(grads, model.trainable_variables))
if step % 200 == 0:
print('Training loss (for one batch) at step %s: %s' % (step, float(loss_value)))
print('Seen so far: %s samples' % ((step + 1) * 64))
#
# Get model
inputs = keras.Input(shape=(784,), name='digits')
x = layers.Dense(64, activation='relu', name='dense_1')(inputs)
x = layers.Dense(64, activation='relu', name='dense_2')(x)
outputs = layers.Dense(10, activation='softmax', name='predictions')(x)
model = keras.Model(inputs=inputs, outputs=outputs)
# Instantiate an optimizer to train the model.
optimizer = keras.optimizers.SGD(learning_rate=1e-3)
# Instantiate a loss function.
loss_fn = keras.losses.SparseCategoricalCrossentropy()
# Prepare the metrics.
train_acc_metric = keras.metrics.SparseCategoricalAccuracy()
val_acc_metric = keras.metrics.SparseCategoricalAccuracy()
# Prepare the training dataset.
batch_size = 64
train_dataset = tf.data.Dataset.from_tensor_slices((x_train, y_train))
train_dataset = train_dataset.shuffle(buffer_size=1024).batch(batch_size)
# Prepare the validation dataset.
val_dataset = tf.data.Dataset.from_tensor_slices((x_val, y_val))
val_dataset = val_dataset.batch(64)
# Iterate over epochs.
for epoch in range(3):
print('Start of epoch %d' % (epoch,))
# Iterate over the batches of the dataset.
for step, (x_batch_train, y_batch_train) in enumerate(train_dataset):
with tf.GradientTape() as tape:
logits = model(x_batch_train)
loss_value = loss_fn(y_batch_train, logits)
grads = tape.gradient(loss_value, model.trainable_variables)
optimizer.apply_gradients(zip(grads, model.trainable_variables))
# Update training metric.
train_acc_metric(y_batch_train, logits)
# Log every 200 batches.
if step % 200 == 0:
print('Training loss (for one batch) at step %s: %s' % (step, float(loss_value)))
print('Seen so far: %s samples' % ((step + 1) * 64))
# Display metrics at the end of each epoch.
train_acc = train_acc_metric.result()
print('Training acc over epoch: %s' % (float(train_acc),))
# Reset training metrics at the end of each epoch
train_acc_metric.reset_states()
# Run a validation loop at the end of each epoch.
for x_batch_val, y_batch_val in val_dataset:
val_logits = model(x_batch_val)
# Update val metrics
val_acc_metric(y_batch_val, val_logits)
val_acc = val_acc_metric.result()
val_acc_metric.reset_states()
print('Validation acc: %s' % (float(val_acc),))
## loss, loss
class ActivityRegularizationLayer(layers.Layer):
def call(self, inputs):
self.add_loss(1e-2 * tf.reduce_sum(inputs))
return inputs
inputs = keras.Input(shape=(784,), name='digits')
x = layers.Dense(64, activation='relu', name='dense_1')(inputs)
# Insert activity regularization as a layer
x = ActivityRegularizationLayer()(x)
x = layers.Dense(64, activation='relu', name='dense_2')(x)
outputs = layers.Dense(10, activation='softmax', name='predictions')(x)
model = keras.Model(inputs=inputs, outputs=outputs)
logits = model(x_train[:64])
print(model.losses)
logits = model(x_train[:64])
logits = model(x_train[64: 128])
logits = model(x_train[128: 192])
print(model.losses)
# loss
optimizer = keras.optimizers.SGD(learning_rate=1e-3)
for epoch in range(3):
print('Start of epoch %d' % (epoch,))
for step, (x_batch_train, y_batch_train) in enumerate(train_dataset):
with tf.GradientTape() as tape:
logits = model(x_batch_train)
loss_value = loss_fn(y_batch_train, logits)
# Add extra losses created during this forward pass:
loss_value += sum(model.losses)
grads = tape.gradient(loss_value, model.trainable_variables)
optimizer.apply_gradients(zip(grads, model.trainable_variables))
# Log every 200 batches.
if step % 200 == 0:
print('Training loss (for one batch) at step %s: %s' % (step, float(loss_value)))
print('Seen so far: %s samples' % ((step + 1) * 64))
Tensor Flow 2.0이keras 트레이닝 모델을 사용하는 실현에 관한 이 글은 여기까지 소개되었습니다. 더 많은 Tensor Flow 2.0keras 트레이닝 모델 내용은 저희 이전의 글을 검색하거나 아래의 관련 글을 계속 훑어보십시오. 앞으로 많은 응원 부탁드립니다!
이 내용에 흥미가 있습니까?
현재 기사가 여러분의 문제를 해결하지 못하는 경우 AI 엔진은 머신러닝 분석(스마트 모델이 방금 만들어져 부정확한 경우가 있을 수 있음)을 통해 가장 유사한 기사를 추천합니다:
KERA의 반복 레이어 시각적 안내서따라서 그림에서 우리는 크기가 4인 숨겨진 상태가 어떻게 RNN 블록 따라서 전체 문장에 대해 우리는 RNN 층에서 크기가 4인 벡터를 출력합니다. 그림에서 보듯이 우리는 무작위 입력 데이터를 만들었는데 그 중 한 ...
텍스트를 자유롭게 공유하거나 복사할 수 있습니다.하지만 이 문서의 URL은 참조 URL로 남겨 두십시오.
CC BY-SA 2.5, CC BY-SA 3.0 및 CC BY-SA 4.0에 따라 라이센스가 부여됩니다.