Facing below Error
ValueError: Exception encountered when calling layer "WeightSharedConvolutionalClassHead" (type WeightSharedConvolutionalClassHead).
In this `tf.Variable` creation, the initial value's shape ((3, 3, 256, 546)) is not compatible with the explicitly supplied `shape` argument ((3, 3, 256, 12)).
Call arguments received:
• features=tf.Tensor(shape=(1, 80, 80, 256), dtype=float32)
This error tells you that the shape of a certain transformation is not as expected, so you need to go back and check why is that!
As a starter my idea is you either are not selecting the right layers or not extending them properly.
@jay_mangi Could you solve it?
kindly make a fresh post even if you similar error, chances are your error could be related to anything else.
post a screenshot of the error you are getting without sharing any code.
You can tag a mentor in case you want them to have look.
Regards
DP