ValueError in model.fit keras and user code

Question

I have a problem in training a model

I'm learning deep-learning python using keras and tensorflow. I am using efficientnetb0 from imagenet dataset. I had divided the training and testing sets and performed one hot encoding. I have 17 folders or classifications of images.

effnet = EfficientNetB0(weights='imagenet',include_top=False,input_shape=(image_size,image_size,3))

model = effnet.output
model = tf.keras.layers.GlobalAveragePooling2D()(model)
model = tf.keras.layers.Dropout(rate=0.5)(model)
model = tf.keras.layers.Dense(17,activation='softmax')(model)
model = tf.keras.models.Model(inputs=effnet.input, outputs = model)

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

Everything ran smooth until training the model

history = model.fit(X_train,y_train,validation_split=0.1, epochs =10, verbose=1,batch_size=32, callbacks=[tensorboard,checkpoint,reduce_lr])

ValueError                                Traceback (most recent call last)
Input In [22], in <cell line: 1>()
----> 1 history = model.fit(X_train,y_train,validation_split=0.1, epochs =20, verbose=1,batch_size=32, callbacks=[tensorboard,checkpoint,reduce_lr])

File C:\Ken\Conda\lib\site-packages\keras\utils\traceback_utils.py:67, in filter_traceback.<locals>.error_handler(*args, **kwargs)
     65 except Exception as e:  # pylint: disable=broad-except
     66   filtered_tb = _process_traceback_frames(e.__traceback__)
---> 67   raise e.with_traceback(filtered_tb) from None
     68 finally:
     69   del filtered_tb

File ~\AppData\Local\Temp\__autograph_generated_filexkmxbaog.py:15, in outer_factory.<locals>.inner_factory.<locals>.tf__train_function(iterator)
     13 try:
     14     do_return = True
---> 15     retval_ = ag__.converted_call(ag__.ld(step_function), (ag__.ld(self), ag__.ld(iterator)), None, fscope)
     16 except:
     17     do_return = False

ValueError: in user code:

    File "C:\Ken\Conda\lib\site-packages\keras\engine\training.py", line 1051, in train_function  *
        return step_function(self, iterator)
    File "C:\Ken\Conda\lib\site-packages\keras\engine\training.py", line 1040, in step_function  **
        outputs = model.distribute_strategy.run(run_step, args=(data,))
    File "C:\Ken\Conda\lib\site-packages\keras\engine\training.py", line 1030, in run_step  **
        outputs = model.train_step(data)
    File "C:\Ken\Conda\lib\site-packages\keras\engine\training.py", line 890, in train_step
        loss = self.compute_loss(x, y, y_pred, sample_weight)
    File "C:\Ken\Conda\lib\site-packages\keras\engine\training.py", line 948, in compute_loss
        return self.compiled_loss(
    File "C:\Ken\Conda\lib\site-packages\keras\engine\compile_utils.py", line 201, in __call__
        loss_value = loss_obj(y_t, y_p, sample_weight=sw)
    File "C:\Ken\Conda\lib\site-packages\keras\losses.py", line 139, in __call__
        losses = call_fn(y_true, y_pred)
    File "C:\Ken\Conda\lib\site-packages\keras\losses.py", line 243, in call  **
        return ag_fn(y_true, y_pred, **self._fn_kwargs)
    File "C:\Ken\Conda\lib\site-packages\keras\losses.py", line 1787, in categorical_crossentropy
        return backend.categorical_crossentropy(
    File "C:\Ken\Conda\lib\site-packages\keras\backend.py", line 5119, in categorical_crossentropy
        target.shape.assert_is_compatible_with(output.shape)

    ValueError: Shapes (None, 18) and (None, 17) are incompatible

Answer 1

I fixed the error

I just changed dense layer to 18 and deleted dropout layer. I don't know why. I will determine if it is because of the dense layer or the dropout layer. Maybe dropout helps the model not to overfit. Maybe it was the dense layer's error.