Logits and label same shape issue

When i am running ‘layer start’ on my image classification for cats_vs_dogs, I am coming across this error:

ValueError: logits and labels must have the same shape ((None, None,
None, 1) vs (None, 1))\n'

Any ideas?

Have tried both of these:

Hi Nisha,

Thank you for reporting this issue, we are looking into it now.

Would it be possible for you to share your layer project and code via GitHub for example?

Best regards,
Adrien

Hi Nisha, I hope you are well.
Ive run a few experiments and also looked at your github code vs your original.
In your github I see you reassign the model several times. Can you describe what you are trying to achieve in this model training please?

The error seems to be this part

model = Sequential()
model.add(Dense(10, activation='relu'))
model.add(Dense(10, activation='sigmoid'))

You probably mean to change the shape before applying Dense ? Or apply it to the existing model instead of a new one?

I hope this helps,
Best regards,
Adrien