summary() function not working in cnn (ValueError: Undefined shapes are not supported.)

Question

I'm trying to make a classification network for IDing pictures from the cifar10 dataset. When I try to use the summary() function, I keep getting this error.

ValueError                                Traceback (most recent call last)
Cell In[267], line 4
      1 #base_model.summary()
      2 #top_model.summary()
      3 #print(base_model.output_shape)
----> 4 model2.summary()

File c:\Users\noahc\anaconda3\Lib\site-packages\keras\src\utils\traceback_utils.py:122, in filter_traceback.<locals>.error_handler(*args, **kwargs)
    119     filtered_tb = _process_traceback_frames(e.__traceback__)
    120     # To get the full stack trace, call:
    121     # `keras.config.disable_traceback_filtering()`
--> 122     raise e.with_traceback(filtered_tb) from None
    123 finally:
    124     del filtered_tb

File c:\Users\noahc\anaconda3\Lib\site-packages\optree\ops.py:747, in tree_map(func, tree, is_leaf, none_is_leaf, namespace, *rests)
    745 leaves, treespec = _C.flatten(tree, is_leaf, none_is_leaf, namespace)
    746 flat_args = [leaves] + [treespec.flatten_up_to(r) for r in rests]
--> 747 return treespec.unflatten(map(func, *flat_args))

ValueError: Undefined shapes are not supported.

Here's the code...

import tensorflow as tf
from keras.applications import VGG16

base_model = VGG16(weights='imagenet', include_top=False, input_shape=(32, 32, 3))

top_model = tf.keras.Sequential([
    layers.Flatten(input_shape=base_model.output_shape[1:]),
    layers.Dense(10, activation='softmax')
])

for layer in base_model.layers[:10]:
    layer.trainable = False

model2 = tf.keras.models.Sequential([
    base_model,
    top_model
])

model2.summary() # Error Occurs Here

I've done a summary for the base and top model and it works fine. But when I test model2 the error occurs. Have no idea why. Not sure what it means by 'undefined' shapes. Not sure what else to try. The summary worked when I only took the first 11 or 15 layers of the vgg16. I've heard it might be a problem with the python version itself, but idk...

Answer 1

I just removed input_shape=base_model.output_shape[1:] and it worked.