una-dinosauria/human-motion-prediction

ValueError: Asked to load checkpoint 10000, but it does not seem to exist

Opened this issue ยท 6 comments

When I run python src/translate.py --action walking --seq_length_out 25 --iterations 10000 --sample --load 10000 after training I get the following error
screen shot 2018-11-08 at 8 43 33 pm

What does the folder experiments/walking/.../not_residual_vel have? The checkpoint should be there.

PS. Please do not post images of text. Just copy the text.

Here's what's inside not_residual_vel
screen shot 2018-11-08 at 9 02 01 pm

Weird. The checkpoint is definitely there. Not sure what is going on -- I suggest you try to figure out why this line

if os.path.isfile(os.path.join(train_dir,"checkpoint-{0}.index".format(FLAGS.load))):

returns False in your machine, since it should return True. Maybe something to do with python's os.path in MacOS?

Another suggestion is to just use linux -- it's free! :)

Thank you for your quick response, there's so many files before not_residual_vel
Is this a path problem?
Here's my path to not_residual_vel

/Users/assam/Desktop/human-motion-prediction/experiments/walking/out_25/iterations_10000/tied/sampling_based/one_hot/depth_1/size_1024/lr_0.005/not_residual_vel

The path looks good to me.

Like I said, you should debug why that line returns False.

Hints: use a debugger; use Google.

I'm sorry I can't help you further since this is pretty specific to your machine and OS.

Cheers,

I fixed the issue by telling the file path.
How can I use the webcam camera as an input?
Thank you