Checkpoint data loss error when evaluating

[chrisyue1998]

I was able to successfully train a model on the stochastic mnist dataset, but when I try to run the evaluation script, I get this error:
tensorflow.python.framework.errors_impl.DataLossError: Unable to open table file logs/smmnist/ours_savp/checkpoint: Data loss: not an sstable (bad magic number): perhaps your file is in a different file format and you need to use a different restoreoperator?
Is there any reason why this is happening? I'm able to successfully evaluate the model pretrained on KTH, but I can't with my own model.

I'm using CUDA 9.0, cuDNN 7.3.0.29, and TF 1.12