Assertion Error in embed_plusrelation_elmo_massspec.py

[nellykan]

Hi! I tried to run the preprocess.py with some data however I get this error:

num of batches 95
  0%|                                                                                                                                                                              | 0/94 [00:00<?, ?it/s]data embedding

running python […]/embed_plusrelation_elmo_massspec.py […arguments…]

Traceback (most recent call last):
  File “./DeepLigand/embed_plusrelation_elmo_massspec.py", line 107, in <module>
    embed_all(f1, f2, f3, f4, f5, args.elmodir, args.elmotag, mapper, args.outfileprefix)
  File “./DeepLigand/embed_plusrelation_elmo_massspec.py", line 52, in embed_all
    assert(exists(join(elmo_dir, 'batch'+str(elmo_cnt)+'.'+elmotag+'.h5py')))
AssertionError

./test.pep.token contains all files from batch1 to batch94 but the files that should have been generated by elmo_embed.py seem to be missing. Do you have any idea why that is?

Thanks!

[haoyangz]

It's hard to tell without more logs. Can you exit preprocess.py at here and check if there is any error from the elmo_embed step?

[nellykan]

I got the assertion error when I tried running preprocess.py with the output to overwrite on an existing folder. I got this Output directory ./DeepLigand/1st_try exists! Overwrite? (yes/no), I typed yes and then I run into the assertion error. However, when I tried to run it specifying as -o a new directory, it ran smoothly.