在运行训练 model 以使用 keras 进行面部表情检测时出现错误

Question

I'm working on the facial expression detection model with python 3.8.8 and TensorFlow 2.4.

My code:

import keras
from keras.preprocessing.image import ImageDataGenerator
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation, Flatten, BatchNormalization
from tensorflow.keras.layers import Conv2D, MaxPooling2D
import os

num_classes = 5
img_rows, img_cols = 48, 48
batch_size = 32

train_data_dir = r'D:\College Projects\SEM6_proj\Music-Recommendation-using-Facial-Expressions-master\train'
validation_data_dir = r'D:\College Projects\SEM6_proj\Music-Recommendation-using-Facial-Expressions-master\validation'

train_datagen = ImageDataGenerator( rescale= 1./255, 
                                    rotation_range= 30, 
                                    shear_range= 0.3, 
                                    zoom_range= 0.3, 
                                    width_shift_range= 0.4, 
                                    height_shift_range= 0.4,
                                    horizontal_flip= True,
                                    vertical_flip= True,
                                    fill_mode= 'nearest')

validation_datagen = ImageDataGenerator(rescale= 1./255)

train_generator = train_datagen.flow_from_directory(
                        train_data_dir,
                        color_mode= 'grayscale',
                        target_size= (img_rows,img_cols),
                        batch_size= batch_size,
                        class_mode= 'categorical',
                        shuffle= True,
                    )

validation_generator = validation_datagen.flow_from_directory(
                        validation_data_dir,
                        color_mode= 'grayscale',
                        target_size= (img_rows,img_cols),
                        batch_size= batch_size,
                        class_mode= 'categorical',
                        shuffle= True,
                    )      


model = Sequential()


# Block - 1          

model.add(Conv2D(32,(3,3),padding = 'same', kernel_initializer = 'he_normal', input_shape = (img_rows,img_cols,1)))    
model.add(Activation('elu'))
model.add(BatchNormalization())
model.add(Conv2D(32,(3,3),padding = 'same', kernel_initializer = 'he_normal', input_shape = (img_rows,img_cols,1)))    
model.add(Activation('elu'))
model.add(BatchNormalization())
model.add(MaxPooling2D(pool_size = (2,2)))
model.add(Dropout(0.2))

# Block - 2

model.add(Conv2D(64,(3,3),padding = 'same', kernel_initializer = 'he_normal', input_shape = (img_rows,img_cols,1)))    
model.add(Activation('elu'))
model.add(BatchNormalization())
model.add(Conv2D(64,(3,3),padding = 'same', kernel_initializer = 'he_normal', input_shape = (img_rows,img_cols,1)))    
model.add(Activation('elu'))
model.add(BatchNormalization())
model.add(MaxPooling2D(pool_size = (2,2)))
model.add(Dropout(0.2))

# Block - 3

model.add(Conv2D(128,(3,3),padding = 'same', kernel_initializer = 'he_normal', input_shape = (img_rows,img_cols,1)))    
model.add(Activation('elu'))
model.add(BatchNormalization())
model.add(Conv2D(128,(3,3),padding = 'same', kernel_initializer = 'he_normal', input_shape = (img_rows,img_cols,1)))    
model.add(Activation('elu'))
model.add(BatchNormalization())
model.add(MaxPooling2D(pool_size = (2,2)))
model.add(Dropout(0.2))

# Block - 4

model.add(Conv2D(256,(3,3),padding = 'same', kernel_initializer = 'he_normal', input_shape = (img_rows,img_cols,1)))    
model.add(Activation('elu'))
model.add(BatchNormalization())
model.add(Conv2D(256,(3,3),padding = 'same', kernel_initializer = 'he_normal', input_shape = (img_rows,img_cols,1)))    
model.add(Activation('elu'))
model.add(BatchNormalization())
model.add(MaxPooling2D(pool_size = (2,2)))
model.add(Dropout(0.2))

# Block - 5

model.add(Flatten())    
model.add(Dense(64, kernel_initializer= 'he_normal'))
model.add(Activation('elu'))
model.add(BatchNormalization())
model.add(Dropout(0.5))

# Block - 6

model.add(Dense(64, kernel_initializer= 'he_normal'))
model.add(Activation('elu'))
model.add(BatchNormalization())
model.add(Dropout(0.5))

# Block - 7

model.add(Dense(num_classes, kernel_initializer= 'he_normal'))
model.add(Activation('softmax'))

# print(model.summary())

from keras.optimizers import RMSprop, SGD, Adam
from keras.callbacks import ModelCheckpoint, EarlyStopping, ReduceLROnPlateau

checkpoint = ModelCheckpoint('D:\College Projects\SEM6_proj\Music-Recommendation-using-Facial-Expressions-master\Emotion_little_vgg.h5',
                              monitor='val_loss',
                              mode ='min',
                              save_best_only= True,
                              verbose=1)

earlystop = EarlyStopping(monitor='val_loss',
                          min_delta= 0,
                          patience=3,
                          verbose=1,
                          restore_best_weights=True)  

reduce_lr = ReduceLROnPlateau(monitor='val_loss',
                              factor= 0.2,
                              patience= 3,
                              verbose=1,
                              min_delta= 0.0001) 

callbacks = [ earlystop, checkpoint, reduce_lr ]

model.compile(loss='categorical_crossentropy',
              optimizer= Adam(lr = 0.001),
              metrics= ['accuracy'])

nb_train_samples = 24176
nb_validation_samples = 3006
epochs = 25

history = model.fit(
                    train_generator,
                    steps_per_epoch= nb_train_samples // batch_size,
                    epochs= epochs,
                    callbacks = callbacks,
                    validation_data= validation_generator,
                    validation_steps= nb_validation_samples // batch_size)

Error:

2021-03-08 18:01:49.114010: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found
2021-03-08 18:01:49.114992: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.        
Found 28821 images belonging to 7 classes.
Found 7066 images belonging to 7 classes.
2021-03-08 18:02:09.707097: I tensorflow/compiler/jit/xla_cpu_device.cc:41] Not creating XLA devices, tf_xla_enable_xla_devices not set
2021-03-08 18:02:09.709548: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'nvcuda.dll'; dlerror: nvcuda.dll not found
2021-03-08 18:02:09.709777: W tensorflow/stream_executor/cuda/cuda_driver.cc:326] failed call to cuInit: UNKNOWN ERROR (303)
2021-03-08 18:02:09.715481: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:169] retrieving CUDA diagnostic information for host: DESKTOP-G870SC4
2021-03-08 18:02:09.715834: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:176] hostname: DESKTOP-G870SC4
2021-03-08 18:02:09.717089: I tensorflow/compiler/jit/xla_gpu_device.cc:99] Not creating XLA devices, tf_xla_enable_xla_devices not set
2021-03-08 18:02:12.330725: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:116] None of the MLIR optimization passes are enabled (registered 2)
Epoch 1/25
Traceback (most recent call last):
  File "D:\College Projects\SEM6_proj\Music-Recommendation-using-Facial-Expressions-master\training.py", line 146, in <module>
    history = model.fit(
  File "D:\College Projects\SEM6_proj\my_venv2\lib\site-packages\tensorflow\python\keras\engine\training.py", line 1100, in fit
    tmp_logs = self.train_function(iterator)
  File "D:\College Projects\SEM6_proj\my_venv2\lib\site-packages\tensorflow\python\eager\def_function.py", line 828, in __call__
    result = self._call(*args, **kwds)
  File "D:\College Projects\SEM6_proj\my_venv2\lib\site-packages\tensorflow\python\eager\def_function.py", line 888, in _call
    return self._stateless_fn(*args, **kwds)
  File "D:\College Projects\SEM6_proj\my_venv2\lib\site-packages\tensorflow\python\eager\function.py", line 2942, in __call__
    return graph_function._call_flat(
  File "D:\College Projects\SEM6_proj\my_venv2\lib\site-packages\tensorflow\python\eager\function.py", line 1918, in _call_flat
    return self._build_call_outputs(self._inference_function.call(
  File "D:\College Projects\SEM6_proj\my_venv2\lib\site-packages\tensorflow\python\eager\function.py", line 555, in call
    outputs = execute.execute(
  File "D:\College Projects\SEM6_proj\my_venv2\lib\site-packages\tensorflow\python\eager\execute.py", line 59, in quick_execute
    tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,
tensorflow.python.framework.errors_impl.InvalidArgumentError:  logits and labels must be broadcastable: logits_size=[32,5] labels_size=[32,7]
         [[node categorical_crossentropy/softmax_cross_entropy_with_logits (defined at \Music-Recommendation-using-Facial-Expressions-master\training.py:146) ]] [Op:__inference_train_function_3059]

Function call stack:
train_function

Answer 1

The error message reports:

logits_size=[32,5] labels_size=[32,7]

ie that your (I guess one-hot encoded) labels have 7 columns; this indeed agrees with the printouts generated by your code above regarding your data:

Found 28821 images belonging to 7 classes.
Found 7066 images belonging to 7 classes.

So, it would certainly seem that your num_classes should be 7, and not 5. Changing to num_classes = 7 should clear the error.

---

As a general rule (and irrelevant to your error here), mixing layers from keras and tensorflow.keras , as you do here, is bad practice and you should not do it; these are actually different packages, so pick one, and take all layers from it - don't mix them.