1、绪论
当在进行有监督学习时,我们可以使用fit()
函数对模型进行训练,通过迭代优化模型的参数,使其能够更好地拟合训练数据。
但当我们希望控制每一个小细节时,就可以完全从头开始编写自己的训练循环。此时就需要一个自定义的训练算法,但是如果我们同时又想受益于fit()
的便捷功能,如回调、内置分布支持或步骤融合,该怎么办呢?
Keras的一个核心原则是复杂性的渐进披露。我们总是能够逐渐进入更底层的工作流程。如果高级功能不完全符合我们的要求,我们就能够通过自定义fit()在保留相应数量高级便利性的同时,对小细节获得更多的控制。
当我们需要自定义fit()
的行为时,你应该重写Model类的训练步骤函数。这是fit()
函数为每一批数据调用的函数。然后,你就可以像平常一样调用fit()
——而它将会运行你自己的学习算法。
2、运行准备
2.1 设置
运行前请按照如下进行设置
import os
# This guide can only be run with the torch backend.
os.environ["KERAS_BACKEND"] = "torch"
import torch
import keras
from keras import layers
import numpy as np
2.2 示例代码
一下我们从一个简单的例子开始感受在PyTorch中自定义fit()函数中的操作的方法。
首先需要创建一个新的类,它继承自keras.Model。
建立这个新类后,只需要重写train_step(self, data)
方法。
运行上述方法将返回一个字典,该字典将指标名称(包括损失)映射到它们的当前值。
输入参数data
是传递给fit
作为训练数据的内容:
- 如果通过调用
fit(x, y, ...)
传递NumPy数组,那么data
将是元组(x, y)
- 如果通过调用
fit(dataset, ...)
传递一个torch.utils.data.DataLoader
或tf.data.Dataset
,那么data
将是数据集在每个批次中生成的内容。
在train_step()
方法的主体中,我们实现了一个常规的训练更新。重要的是,我们通过self.compute_loss()
计算损失,该方法封装了在compile()
方法中传递的损失函数。
类似地,我们对self.metrics
中的指标调用metric.update_state(y, y_pred)
,以更新在compile()
方法中传递的指标的状态,并在最后查询self.metrics
以检索它们的当前值。
class CustomModel(keras.Model):
def train_step(self, data):
# Unpack the data. Its structure depends on your model and
# on what you pass to `fit()`.
x, y = data
# Call torch.nn.Module.zero_grad() to clear the leftover gradients
# for the weights from the previous train step.
self.zero_grad()
# Compute loss
y_pred = self(x, training=True) # Forward pass
loss = self.compute_loss(y=y, y_pred=y_pred)
# Call torch.Tensor.backward() on the loss to compute gradients
# for the weights.
loss.backward()
trainable_weights = [v for v in self.trainable_weights]
gradients = [v.value.grad for v in trainable_weights]
# Update weights
with torch.no_grad():
self.optimizer.apply(gradients, trainable_weights)
# Update metrics (includes the metric that tracks the loss)
for metric in self.metrics:
if metric.name == "loss":
metric.update_state(loss)
else:
metric.update_state(y, y_pred)
# Return a dict mapping metric names to current value
# Note that it will include the loss (tracked in self.metrics).
return {m.name: m.result() for m in self.metrics}
运行代码,输出如下所示
# Construct and compile an instance of CustomModel
inputs = keras.Input(shape=(32,))
outputs = keras.layers.Dense(1)(inputs)
model = CustomModel(inputs, outputs)
model.compile(optimizer="adam", loss="mse", metrics=["mae"])
# Just use `fit` as usual
x = np.random.random((1000, 32))
y = np.random.random((1000, 1))
model.fit(x, y, epochs=3)
Epoch 1/3
32/32 ━━━━━━━━━━━━━━━━━━━━ 0s 551us/step - mae: 0.6533 - loss: 0.6036
Epoch 2/3
32/32 ━━━━━━━━━━━━━━━━━━━━ 0s 522us/step - mae: 0.4013 - loss: 0.2522
Epoch 3/3
32/32 ━━━━━━━━━━━━━━━━━━━━ 0s 516us/step - mae: 0.3813 - loss: 0.2256
<keras.src.callbacks.history.History at 0x299b7baf0>
3、底层操作
当然,实际操作过程中也可以在compile()
方法中不传递损失函数,而是在train_step
中手动处理所有事情。同样地,对于指标也是如此。
下面是一个更底层级别操作的例子,它仅使用compile()
来配置优化器:
我们首先创建Metric
实例来跟踪我们的损失和MAE分数(在__init__()
方法中)。
通过一个自定义的train_step()
,更新这些指标的状态(通过在其上调用update_state()
),然后查询它们(通过result()
)以返回它们的当前平均值,这些值将由进度条显示并传递给任何回调。
请注意,运行过程中需要在每个epoch之间调用reset_states()
来重置指标!否则,调用result()
将返回从训练开始以来的平均值,而通常是使用每个epoch的平均值。框架可以为我们做这件事:只需将你想要重置的任何指标列在模型的metrics
属性中即可。模型将在每个fit()
epoch的开始或evaluate()
调用的开始时调用reset_states()
来重置这些对象的状态。
class CustomModel(keras.Model):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.loss_tracker = keras.metrics.Mean(name="loss")
self.mae_metric = keras.metrics.MeanAbsoluteError(name="mae")
self.loss_fn = keras.losses.MeanSquaredError()
def train_step(self, data):
x, y = data
# Call torch.nn.Module.zero_grad() to clear the leftover gradients
# for the weights from the previous train step.
self.zero_grad()
# Compute loss
y_pred = self(x, training=True) # Forward pass
loss = self.loss_fn(y, y_pred)
# Call torch.Tensor.backward() on the loss to compute gradients
# for the weights.
loss.backward()
trainable_weights = [v for v in self.trainable_weights]
gradients = [v.value.grad for v in trainable_weights]
# Update weights
with torch.no_grad():
self.optimizer.apply(gradients, trainable_weights)
# Compute our own metrics
self.loss_tracker.update_state(loss)
self.mae_metric.update_state(y, y_pred)
return {
"loss": self.loss_tracker.result(),
"mae": self.mae_metric.result(),
}
@property
def metrics(self):
# We list our `Metric` objects here so that `reset_states()` can be
# called automatically at the start of each epoch
# or at the start of `evaluate()`.
return [self.loss_tracker, self.mae_metric]
# Construct an instance of CustomModel
inputs = keras.Input(shape=(32,))
outputs = keras.layers.Dense(1)(inputs)
model = CustomModel(inputs, outputs)
# We don't pass a loss or metrics here.
model.compile(optimizer="adam")
# Just use `fit` as usual -- you can use callbacks, etc.
x = np.random.random((1000, 32))
y = np.random.random((1000, 1))
model.fit(x, y, epochs=5)
Epoch 1/5
32/32 ━━━━━━━━━━━━━━━━━━━━ 0s 461us/step - loss: 0.2470 - mae: 0.3953
Epoch 2/5
32/32 ━━━━━━━━━━━━━━━━━━━━ 0s 456us/step - loss: 0.2386 - mae: 0.3910
Epoch 3/5
32/32 ━━━━━━━━━━━━━━━━━━━━ 0s 456us/step - loss: 0.2359 - mae: 0.3901
Epoch 4/5
32/32 ━━━━━━━━━━━━━━━━━━━━ 0s 480us/step - loss: 0.2013 - mae: 0.3572
Epoch 5/5
32/32 ━━━━━━━━━━━━━━━━━━━━ 0s 463us/step - loss: 0.1903 - mae: 0.3480
<keras.src.callbacks.history.History at 0x299c5eec0>
3.1 支持样本权重和分类权重
在文章开始的基本示例没有提到样本权重,那么如果想要支持fit()
方法的sample_weight
和class_weight
参数,可以按照以下步骤进行:
从data
参数中解包sample_weight
将其传递给compute_loss
和update_state
(当然,如果你不是依赖于compile()
方法来设置损失和指标,也可以手动应用它)
class CustomModel(keras.Model):
def train_step(self, data):
# Unpack the data. Its structure depends on your model and
# on what you pass to `fit()`.
if len(data) == 3:
x, y, sample_weight = data
else:
sample_weight = None
x, y = data
# Call torch.nn.Module.zero_grad() to clear the leftover gradients
# for the weights from the previous train step.
self.zero_grad()
# Compute loss
y_pred = self(x, training=True) # Forward pass
loss = self.compute_loss(
y=y,
y_pred=y_pred,
sample_weight=sample_weight,
)
# Call torch.Tensor.backward() on the loss to compute gradients
# for the weights.
loss.backward()
trainable_weights = [v for v in self.trainable_weights]
gradients = [v.value.grad for v in trainable_weights]
# Update weights
with torch.no_grad():
self.optimizer.apply(gradients, trainable_weights)
# Update metrics (includes the metric that tracks the loss)
for metric in self.metrics:
if metric.name == "loss":
metric.update_state(loss)
else:
metric.update_state(y, y_pred, sample_weight=sample_weight)
# Return a dict mapping metric names to current value
# Note that it will include the loss (tracked in self.metrics).
return {m.name: m.result() for m in self.metrics}
# Construct and compile an instance of CustomModel
inputs = keras.Input(shape=(32,))
outputs = keras.layers.Dense(1)(inputs)
model = CustomModel(inputs, outputs)
model.compile(optimizer="adam", loss="mse", metrics=["mae"])
# You can now use sample_weight argument
x = np.random.random((1000, 32))
y = np.random.random((1000, 1))
sw = np.random.random((1000, 1))
model.fit(x, y, sample_weight=sw, epochs=3)
Epoch 1/3
32/32 ━━━━━━━━━━━━━━━━━━━━ 0s 499us/step - mae: 1.4332 - loss: 1.0769
Epoch 2/3
32/32 ━━━━━━━━━━━━━━━━━━━━ 0s 520us/step - mae: 0.9250 - loss: 0.5614
Epoch 3/3
32/32 ━━━━━━━━━━━━━━━━━━━━ 0s 502us/step - mae: 0.6069 - loss: 0.2653
<keras.src.callbacks.history.History at 0x299c82bf0>
3.2 提供自定义的评估步骤
如果想要在调用model.evaluate()
时自定义评估步骤,我们怎么做呢?那么我们将以完全相同的方式重写test_step
。
class CustomModel(keras.Model):
def test_step(self, data):
# Unpack the data
x, y = data
# Compute predictions
y_pred = self(x, training=False)
# Updates the metrics tracking the loss
loss = self.compute_loss(y=y, y_pred=y_pred)
# Update the metrics.
for metric in self.metrics:
if metric.name == "loss":
metric.update_state(loss)
else:
metric.update_state(y, y_pred)
# Return a dict mapping metric names to current value.
# Note that it will include the loss (tracked in self.metrics).
return {m.name: m.result() for m in self.metrics}
# Construct an instance of CustomModel
inputs = keras.Input(shape=(32,))
outputs = keras.layers.Dense(1)(inputs)
model = CustomModel(inputs, outputs)
model.compile(loss="mse", metrics=["mae"])
# Evaluate with our custom test_step
x = np.random.random((1000, 32))
y = np.random.random((1000, 1))
model.evaluate(x, y)
32/32 ━━━━━━━━━━━━━━━━━━━━ 0s 325us/step - mae: 0.4427 - loss: 0.2993
[0.2726495862007141, 0.42286917567253113]
4、完整的应用示例
为了整合前面所学的知识,我们将通过一个端到端的GAN(生成对抗网络)示例来演示在PyTorch中自定义fit()函数中的操作。
在这个例子中,我们将考虑:
- 一个用于生成28x28x1图像的生成器网络。
- 一个用于将28x28x1图像分类为两个类别(“假”和“真”)的判别器网络。
- 每个网络都有一个优化器。
- 一个用于训练判别器的损失函数。
首先,我们需要定义生成器和判别器的网络结构。这里为了简洁,我们不会详细写出每个层的定义,但你可以想象生成器网络将噪声作为输入并输出图像,而判别器网络将图像作为输入并输出一个概率值,该值表示输入图像是真实的(来自训练集)还是假的(由生成器生成)。
下面是GAN训练的大致流程:
-
初始化生成器和判别器网络:
- 定义生成器和判别器的模型结构。
- 编译判别器网络,并指定一个损失函数(如二元交叉熵)和优化器(如Adam)。
-
训练判别器:
- 对于一批真实图像,计算判别器的损失(使用真实标签1)。
- 通过生成器生成一批假图像,并计算判别器对假图像的损失(使用假标签0)。
- 将两个损失相加,并对判别器执行一次梯度下降更新。
-
训练生成器:
- 生成一批假图像。
- 使用判别器对这些假图像进行预测,得到概率值。
- 使用判别器的预测作为标签(我们想要生成器生成的图像被判别器认为是真实的),计算生成器的损失(这通常是通过将判别器的预测传递给某种损失函数,如二元交叉熵或均方误差,来实现的)。
- 使用计算出的损失对生成器执行一次梯度下降更新。
- 注意:在训练生成器时,我们需要将判别器的权重设置为不可训练(因为我们只希望更新生成器的权重)。这可以通过在训练生成器之前调用
discriminator.trainable = False
来实现。
-
循环迭代:
- 重复步骤2和3多次,以训练GAN。
-
在测试集上评估GAN:
- 使用训练好的生成器生成图像,并可视化这些图像以评估GAN的性能。
# Create the discriminator
discriminator = keras.Sequential(
[
keras.Input(shape=(28, 28, 1)),
layers.Conv2D(64, (3, 3), strides=(2, 2), padding="same"),
layers.LeakyReLU(negative_slope=0.2),
layers.Conv2D(128, (3, 3), strides=(2, 2), padding="same"),
layers.LeakyReLU(negative_slope=0.2),
layers.GlobalMaxPooling2D(),
layers.Dense(1),
],
name="discriminator",
)
# Create the generator
latent_dim = 128
generator = keras.Sequential(
[
keras.Input(shape=(latent_dim,)),
# We want to generate 128 coefficients to reshape into a 7x7x128 map
layers.Dense(7 * 7 * 128),
layers.LeakyReLU(negative_slope=0.2),
layers.Reshape((7, 7, 128)),
layers.Conv2DTranspose(128, (4, 4), strides=(2, 2), padding="same"),
layers.LeakyReLU(negative_slope=0.2),
layers.Conv2DTranspose(128, (4, 4), strides=(2, 2), padding="same"),
layers.LeakyReLU(negative_slope=0.2),
layers.Conv2D(1, (7, 7), padding="same", activation="sigmoid"),
],
name="generator",
)
下面是一个功能完整的GAN类,它重写了compile()
方法以使用自己的签名,并在train_step
中以17行代码实现了整个GAN算法:
class GAN(keras.Model):
def __init__(self, discriminator, generator, latent_dim):
super().__init__()
self.discriminator = discriminator
self.generator = generator
self.latent_dim = latent_dim
self.d_loss_tracker = keras.metrics.Mean(name="d_loss")
self.g_loss_tracker = keras.metrics.Mean(name="g_loss")
self.seed_generator = keras.random.SeedGenerator(1337)
self.built = True
@property
def metrics(self):
return [self.d_loss_tracker, self.g_loss_tracker]
def compile(self, d_optimizer, g_optimizer, loss_fn):
super().compile()
self.d_optimizer = d_optimizer
self.g_optimizer = g_optimizer
self.loss_fn = loss_fn
def train_step(self, real_images):
device = "cuda" if torch.cuda.is_available() else "cpu"
if isinstance(real_images, tuple):
real_images = real_images[0]
# Sample random points in the latent space
batch_size = real_images.shape[0]
random_latent_vectors = keras.random.normal(
shape=(batch_size, self.latent_dim), seed=self.seed_generator
)
# Decode them to fake images
generated_images = self.generator(random_latent_vectors)
# Combine them with real images
real_images = torch.tensor(real_images, device=device)
combined_images = torch.concat([generated_images, real_images], axis=0)
# Assemble labels discriminating real from fake images
labels = torch.concat(
[
torch.ones((batch_size, 1), device=device),
torch.zeros((batch_size, 1), device=device),
],
axis=0,
)
# Add random noise to the labels - important trick!
labels += 0.05 * keras.random.uniform(labels.shape, seed=self.seed_generator)
# Train the discriminator
self.zero_grad()
predictions = self.discriminator(combined_images)
d_loss = self.loss_fn(labels, predictions)
d_loss.backward()
grads = [v.value.grad for v in self.discriminator.trainable_weights]
with torch.no_grad():
self.d_optimizer.apply(grads, self.discriminator.trainable_weights)
# Sample random points in the latent space
random_latent_vectors = keras.random.normal(
shape=(batch_size, self.latent_dim), seed=self.seed_generator
)
# Assemble labels that say "all real images"
misleading_labels = torch.zeros((batch_size, 1), device=device)
# Train the generator (note that we should *not* update the weights
# of the discriminator)!
self.zero_grad()
predictions = self.discriminator(self.generator(random_latent_vectors))
g_loss = self.loss_fn(misleading_labels, predictions)
grads = g_loss.backward()
grads = [v.value.grad for v in self.generator.trainable_weights]
with torch.no_grad():
self.g_optimizer.apply(grads, self.generator.trainable_weights)
# Update metrics and return their value.
self.d_loss_tracker.update_state(d_loss)
self.g_loss_tracker.update_state(g_loss)
return {
"d_loss": self.d_loss_tracker.result(),
"g_loss": self.g_loss_tracker.result(),
}
以下是运行结果
# Prepare the dataset. We use both the training & test MNIST digits.
batch_size = 64
(x_train, _), (x_test, _) = keras.datasets.mnist.load_data()
all_digits = np.concatenate([x_train, x_test])
all_digits = all_digits.astype("float32") / 255.0
all_digits = np.reshape(all_digits, (-1, 28, 28, 1))
# Create a TensorDataset
dataset = torch.utils.data.TensorDataset(
torch.from_numpy(all_digits), torch.from_numpy(all_digits)
)
# Create a DataLoader
dataloader = torch.utils.data.DataLoader(dataset, batch_size=batch_size, shuffle=True)
gan = GAN(discriminator=discriminator, generator=generator, latent_dim=latent_dim)
gan.compile(
d_optimizer=keras.optimizers.Adam(learning_rate=0.0003),
g_optimizer=keras.optimizers.Adam(learning_rate=0.0003),
loss_fn=keras.losses.BinaryCrossentropy(from_logits=True),
)
gan.fit(dataloader, epochs=1)
1094/1094 ━━━━━━━━━━━━━━━━━━━━ 1582s 1s/step - d_loss: 0.3581 - g_loss: 2.0571
<keras.src.callbacks.history.History at 0x299ce1840>