ValueError about pepnet HOT 4 OPEN

https://zenodo.org/record/7869847 , also listed in readme

from pepnet.

Did you use the correct corresponding version of the model files? I've updated this repository several times

from pepnet.

I downloaded the code from github, where can I get the last version?
thanks!

from pepnet.

Hello,
the de novo algorithm works pretty well on the model.h5 released by you using the following command.

python denovo.py --input example.mgf --model model.h5 --output example_prediction.tsv

However, when I trained model on different datasets or even the train.py and validation.py datasets posted by you, it created a model that doesn't match with the post processing function. Here is the bug:
` predictions = model.predict(data_seq(spectra, input_processor, batch_size, xonly=True), verbose=1)

for rst, sp in zip(predictions, spectra):
    ms, c = sp['mass'], sp['charge']

    # run post correction
    pep, pos, positional_score = post_correction(rst, ms, c)`

it returns this error:
Traceback (most recent call last):
File "denovo.py", line 229, in
peps, ppeps, scores, pscores, ppms, _ = denovo(model, spectra, args.batch_size)
File "denovo.py", line 182, in denovo
pep, pos, positional_score = post_correction(rst, ms, c)
File "denovo.py", line 73, in post_correction
pep = topep(seq)
File "/PepNet-master/utils.py", line 169, in topep
return ''.join(map(lambda n: id2amino[n], seq)).strip("[]")
File "/PepNet-master/utils.py", line 169, in
return ''.join(map(lambda n: id2amino[n], seq)).strip("[]")
KeyError: 121

Because the prediction is a list with length of 8 and with eight different shape: (124, 1)
prediction shape:
(124, 32)
(124, 1)
(124, 8)
(124, 21)
(124, 21, 30)
(124, 400)
(124, 32, 24)
124 is the number of spectra in the train.py
Could you please let me know how I can fix it?
Thanks!

from pepnet.