cant load checkpoint files

[skywo1f]

tried all of the files in the checkpoints folder:
model.ckpt.index
model.ckpt.meta
checkpoint
model.ckpt.data-00000-of-00001
none of them work:
Semantic-Segmentation-Suite/checkpoints/0295/model.ckpt.data-00000-of-00001: Data loss: not an sstable (bad magic number): perhaps your file is in a different file format and you need to use a different restore operator

[sweetboxwwy]

Have you solve this problem?

[AI-ML-Enthusiast]

@skywo1f @GeorgeSeif
same problem. Anyone suggest me please.
I trained on my PC , training is Ok but can not load the checkpoint file

[millermuttu]

same issue -

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 BatchNorm/beta not found in checkpoint
[[Node: save/RestoreV2 = RestoreV2[dtypes=[DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, ..., DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT, DT_FLOAT], _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_save/Const_0_0, save/RestoreV2/tensor_names, save/RestoreV2/shape_and_slices)]]
[[Node: save/RestoreV2/_79 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device_incarnation=1, tensor_name="edge_84_save/RestoreV2", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:GPU:0"]]

[FZY2019]

Is the problem solved？

[skywo1f]

no

[Imperssonator]

Had the same issue - in your case, you should just pass model.ckpt
https://stackoverflow.com/questions/33759623/tensorflow-how-to-save-restore-a-model
Also if you're not using CamVid, make sure to pass something to --dataset, otherwise it will default to the 32 class labels from the CamVid dataset. Hope that helps.

[wy9884255]

you better try this disk:/your_folder/model.ckpt

[Harikrishnan24]

Is this problem solved

[mtylerpreston]

I think I have the solution...just use model.ckpt even if no such file exists.

I had the same struggle in trying to pass model.ckpt.meta etc in for resuming the train. Even though no file name exists in the directory where the checkpoint was specified to be saved, I just used model.ckpt and it worked out.

[Harikrishnan24]

Thank you for your help

…

On Fri, May 15, 2020, 10:59 PM M. Tyler Preston ***@***.***> wrote: I had the same struggle. Even though no file name exists in the directory where the checkpoint was specified to be saved, I just used model.ckpt and it worked out. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#225 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEVXEBDB5VG3K656S435TWTRRV3YFANCNFSM4JANDEMQ> .