gcn icon indicating copy to clipboard operation
gcn copied to clipboard

Problems loading a pre-trained model

Open wolfiex opened this issue 5 years ago • 1 comments

Hi, I am trying to save the model with model.save(sess) at the end of the session, and then reload it when supplying additional training data at a later point. This works fine in an interactive session, where saving and loading are directly after eachother.

However in trying to do this for a new program run the following is incurred:

NotFoundError (see above for traceback): Restoring from checkpoint failed. This is most likely due to a Variable name or other graph key that is missing from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error:

Key gcn/graphconvolution_65_vars/weights_0:0 not found in checkpoint
	 [[node save_1/RestoreV2 (defined at /work/home/.../gcn/gcn/models.py:79) ]]

wolfiex avatar Aug 28 '19 05:08 wolfiex

would you mind sharing your pretrained model? @tkipf

Ai-is-light avatar Sep 05 '19 08:09 Ai-is-light