tf.data 多输出模型具有形状不兼容的标签

Posted 2023-02-16

【中文标题】tf.data 多输出模型具有形状不兼容的标签

【英文标题】：tf.data multi output model has labels with incompatible shapes

【发布时间】：2021-12-22 00:01:24

【问题描述】：

我正在尝试将我之前在 Colab 上（使用 ImageDataGenerator）所做的工作簿转换为使用 tf.data.dataset 的工作簿，因为我现在设置了多 GPU，并且正在尝试学习如何进行更快的训练.该模型在来自 Kaggle 的年龄/性别/种族数据集上进行训练，但在这种情况下，我们只对性别和年龄预测感兴趣。 Sex 为 0 或 1，损失函数为二元交叉熵，age 为 0 到 120 之间的整数，损失函数为 mse，表示回归。

import tensorflow as tf
import os

AUTOTUNE = tf.data.AUTOTUNE
batch_size = 64

#Load datasets from directories
train_gen = tf.data.Dataset.list_files(os.listdir(training_dir), shuffle = False)
valid_gen = tf.data.Dataset.list_files(os.listdir(validation_dir), shuffle = False)

def decode_img(img):
    #Convert compressed string into a 3D tensor
    img = tf.io.decode_jpeg(img, channels=3)
    img = tf.image.convert_image_dtype(img, tf.float32)
    #Resize the image to the desired size
    return tf.image.resize(img, [128,128])


def get_label(file):
    
    gender = get_sex(file) #returns either 0 or 1
    age = get_age(file)     #returns interger between 0 and about 120
    return gender, age

def process_path(file):
    file = file.numpy()
    file_path = str(bytes.decode(file))
    file = file_path.split(' ')[-1].split("\\")[-1]
    labels = get_label(file)
    # Load data from file as a String
    img = tf.io.read_file(file_path)
    img = decode_img(img)
    img = img / 255.0

    return img, labels


def _set_shapes(t1, t2):
    t1.set_shape((128,128,3))
    t2.set_shape((2,))
        
    return (t1,t2)


train_gen = train_gen.map(lambda x: tf.py_function(process_path, [x], [tf.float32, tf.int32]), num_parallel_calls=AUTOTUNE)
valid_gen = valid_gen.map(lambda x: tf.py_function(process_path, [x], [tf.float32, tf.int32]), num_parallel_calls=AUTOTUNE)

train_gen = train_gen.map(_set_shapes,num_parallel_calls=AUTOTUNE)
valid_gen = valid_gen.map(_set_shapes, num_parallel_calls=AUTOTUNE)


train_gen = train_gen.batch(batch_size)
valid_gen = valid_gen.batch(batch_size)

train_gen

输出：

#configure for performance
def config_for_performance(ds):
    ds = ds.cache()
    ds = ds.prefetch(buffer_size=AUTOTUNE)
    return ds

train_gen = config_for_performance(train_gen)
valid_gen = config_for_performance(valid_gen)

模型本身：

from tensorflow.keras.models import Model
from tensorflow.keras.layers import Conv2D, MaxPool2D, Dense, Dropout, Input, Activation, Flatten, BatchNormalization, PReLU
from tensorflow.keras.regularizers import l2
from tensorflow.keras.losses import BinaryCrossentropy
from keras.callbacks import EarlyStopping, ReduceLROnPlateau
from tensorflow.keras import mixed_precision

mixed_precision.set_global_policy('mixed_float16')

gpus = tf.config.list_logical_devices('GPU')
#print(gpus)
strategy = tf.distribute.MirroredStrategy(gpus,cross_device_ops=tf.distribute.ReductionToOneDevice())
with strategy.scope():

    #Define the convolution layers
    inp = Input(shape=(128,128,3))
    cl1 = Conv2D(32,(3,3), padding='same', kernel_regularizer=l2(0.001), kernel_initializer='he_uniform')(inp)
    bn1 = BatchNormalization()(cl1)
    pr1 = PReLU(alpha_initializer='he_uniform')(bn1)
    cl2 = Conv2D(32,(3,3), padding='same',kernel_regularizer=l2(0.001), kernel_initializer='he_uniform')(pr1)
    bn2 = BatchNormalization()(cl2)
    pr2 = PReLU(alpha_initializer='he_uniform')(bn2)
    mp1 = MaxPool2D((2,2))(pr2)
    cl3 = Conv2D(64,(3,3), padding='same',kernel_regularizer=l2(0.001), kernel_initializer='he_uniform')(mp1)
    bn3 = BatchNormalization()(cl3)
    pr3 = PReLU(alpha_initializer='he_uniform')(bn3)
    cl4 = Conv2D(64,(3,3), padding='same',kernel_regularizer=l2(0.001), kernel_initializer='he_uniform')(pr3)
    bn4 = BatchNormalization()(cl4)
    pr4 = PReLU(alpha_initializer='he_uniform')(bn4)
    mp2 = MaxPool2D((2,2))(pr4)
    cl5 = Conv2D(128,(3,3), padding='same',kernel_regularizer=l2(0.001), kernel_initializer='he_uniform')(mp2)
    bn5 = BatchNormalization()(cl5)
    pr5 = PReLU(alpha_initializer='he_uniform')(bn5)
    mp3 = MaxPool2D((2,2))(pr5)
    cl6 = Conv2D(256,(3,3), padding='same',kernel_regularizer=l2(0.001), kernel_initializer='he_uniform')(mp3)
    bn6 = BatchNormalization()(cl6)
    pr6 = PReLU(alpha_initializer='he_uniform')(bn6)
    mp4 = MaxPool2D((2,2))(pr6)
    cl7 = Conv2D(512,(3,3), padding='same',kernel_regularizer=l2(0.001), kernel_initializer='he_uniform')(mp4)
    bn7 = BatchNormalization()(cl7)
    pr7 = PReLU(alpha_initializer='he_uniform')(bn7)
    mp5 = MaxPool2D((2,2))(pr7)
    flt = Flatten()(mp5)

    #This layer predicts age
    agelayer = Dense(128, activation='relu',kernel_regularizer=l2(0.001), kernel_initializer='he_uniform')(flt)
    agelayer = BatchNormalization()(agelayer)
    agelayer = Dropout(0.6)(agelayer)
    agelayer = Dense(1, activation='relu', name='age_output', kernel_initializer='he_uniform', dtype='float32')(agelayer)

    #This layer predicts gender
    glayer = Dense(128, activation='relu',kernel_regularizer=l2(0.001), kernel_initializer='he_uniform')(flt)
    glayer = BatchNormalization()(glayer)
    glayer = Dropout(0.5)(glayer)
    glayer = Dense(1, activation='sigmoid', name='gender_output', kernel_initializer='he_uniform', dtype='float32')(glayer)

    modelA = Model(inputs=inp, outputs=[glayer,agelayer])

    model_folder = 'C:/Users/mm/OneDrive/Documents/Age estimation & gender classification/models'

    if not os.path.exists(model_folder):
        os.mkdir(model_folder)

    #Callback to control learning rate during training. Reduces learning rate by 5% after 3 epochs of no improvement on validation loss
    lr_callback = ReduceLROnPlateau(monitor='val_loss', factor=0.95, patience=3,min_lr=0.000005)

    #Callback to stop training if after 100 epochs of no improvement it stops and restores the best weights
    es_callback = EarlyStopping(monitor='val_loss', patience=100, restore_best_weights=True, min_delta=0.001)

    #Compile Model A
    modelA.compile(optimizer='Adam', loss='gender_output': BinaryCrossentropy(), 'age_output': 'mse', metrics='gender_output': 'accuracy', 'age_output':'mae')

#Training Model A
history = modelA.fit(train_gen, epochs=100, validation_data=valid_gen, callbacks=[es_callback,lr_callback])

错误信息：

INFO:tensorflow:在设备上使用 MirroredStrategy ('/job:localhost/replica:0/task:0/device:GPU:0', '/job:localhost/replica:0/task:0/device:GPU :1') 纪元 1/100 INFO:tensorflow:Error 报告给 Coordinator：logits 和标签必须具有相同的形状 ((None, 1) vs (None, 2)) 回溯（最近一次通话最后）： 文件“C:\Users\mm\AppData\Roaming\Python\Python39\site-packages\tensorflow\python\ops\nn_impl.py”，第 130 行，sigmoid_cross_entropy_with_logits 标签.get_shape().assert_is_compatible_with(logits.get_shape()) 文件“C:\Users\mm\AppData\Roaming\Python\Python39\site-packages\tensorflow\python\framework\tensor_shape.py”，第 1161 行，在 assert_is_compatible_with raise ValueError("形状 %s 和 %s 不兼容" % (self, other)) ValueError: Shapes (None, 2) 和 (None, 1) 不兼容

在处理上述异常的过程中，又发生了一个异常：

Traceback（最近一次调用最后一次）： 文件“C:\Users\mm\AppData\Roaming\Python\Python39\site-packages\tensorflow\python\training\coordinator.py”，第 297 行，在 stop_on_exception 屈服 运行中的文件“C:\Users\mm\AppData\Roaming\Python\Python39\site-packages\tensorflow\python\distribute\mirrored_run.py”，第 346 行 self.main_result = self.main_fn(*self.main_args, **self.main_kwargs) 文件“C:\Users\mm\AppData\Roaming\Python\Python39\site-packages\tensorflow\python\autograph\impl\api.py”，第 692 行，在包装器中 return convert_call(f, args, kwargs, options=options) 文件“C:\Users\mm\AppData\Roaming\Python\Python39\site-packages\tensorflow\python\autograph\impl\api.py”，第 382 行，在 convert_call 中 return _call_unconverted(f, args, kwargs, 选项) _call_unconverted 中的文件“C:\Users\mm\AppData\Roaming\Python\Python39\site-packages\tensorflow\python\autograph\impl\api.py”，第 463 行 返回 f(*args, **kwargs) 文件“C:\Users\mm\AppData\Roaming\Python\Python39\site-packages\keras\engine\training.py”，第 835 行，在 run_step 输出 = model.train_step（数据） 显示更多（在文本编辑器中打开原始输出数据）...

文件“C:\Users\mm\AppData\Roaming\Python\Python39\site-packages\tensorflow\python\util\dispatch.py​​”，第 206 行，在包装器中 返回目标（*args，**kwargs） 文件“C:\Users\mm\AppData\Roaming\Python\Python39\site-packages\tensorflow\python\ops\nn_impl.py”，第 132 行，sigmoid_cross_entropy_with_logits raise ValueError("logits 和标签必须具有相同的形状 (%s vs %s)" % ValueError: logits 和 labels 必须具有相同的形状 ((None, 1) vs (None, 2))

【问题讨论】：

请修剪您的代码，以便更容易找到您的问题。请按照以下指南创建minimal reproducible example。

【参考方案1】：

通过一些研究和反复试验设法解决了这个问题。主要问题是：

标签作为元组而不是被分开提供给模型。当它是多个输出头时，这是必要的：

def process_path(file):
    file = file.numpy()
    file_path = str(bytes.decode(file))
    file = file_path.split("\\")[-1]
    gender, age = get_label(file)
    # Load data from file as a String
    img = tf.io.read_file(file_path)
    img = decode_img(img)
    img = img / 255.0

    return img, gender, age

注意：我对从文件名中提取标签进行了修改，因为它并非一直都正确：

file = file_path.split("\\")[-1]

由于1上的变化，map函数需要另外一个标签的dtype，所以变成：

train_gen = train_gen.map(lambda x: tf.py_function(process_path, [x], [tf.float32, tf.int32, tf.int32]), num_parallel_calls=AUTOTUNE)
valid_gen = valid_gen.map(lambda x: tf.py_function(process_path, [x], [tf.float32, tf.int32, tf.int32]), num_parallel_calls=AUTOTUNE)

每个标签都需要重塑：

def _set_shapes(t1, t2, t3):
    t1.set_shape((128,128,3))
    t2.set_shape((1,))
    t3.set_shape((1,))
    t2 = tf.reshape(t2, [-1,1])
    t3 = tf.reshape(t3, [-1,1])
    return (t1,t2,t3)