lkytal / pepnet Goto Github PK

After running the following command:
python denovo.py --input example.mgf --model model.h5 --output example_prediction.tsv
I get this error
lib/python3.7/site-packages/keras/engine/input_spec.py", line 200, in assert_input_compatibility
raise ValueError(f'Layer "{layer_name}" expects {len(input_spec)} input(s),'

V**alueError: Layer "denovo" expects 4 input(s), but it received 3 input tensors. Inputs received**: [<tf.Tensor 'IteratorGetNext:0' shape=(None, None, None) dtype=float32>, <tf.Tensor 'IteratorGetNext:1' shape=(None, None) dtype=float32>, <tf.Tensor 'IteratorGetNext:2' shape=(None, None) dtype=float32>]

I couldn't figure out how to fix it, could you please let me know how I can fix it?
I am using the same inputs and pretrained model provided by the author.
Thanks!

Dear authors,

Thank you for this nice tool. The pre-print sounds promissing and I wanted to test it today. During the installation I noticed, that the requirements in readme.md seem to not be complete: besides tensorflow there is also the need for tensorflow_addons

Could you please add this information?

Best regards,
Juergen

Hi,

Our dataset have lots of modifications which are important, is it possible to modify the model a bit, such that we can train PepNet from scratch by ourselves to be able to include modifications? Otherwise we are still unable to use PepNet for our work.

Best regards

Hi,

I came across a paper related to PepNet. I want to integrate it in my de novo sequencing pipeline, but I am wondering if it is possible to get the top N best candidate peptides for each spectrum.

Best,

Carlos

I'm currently considering including PepNet in Galaxy, i.e. to add a Galaxy tool. For this, a conda package for Pepnet would be required. I could work on both (actually I already started: bioconda/bioconda-recipes#42751)

I have a few questions or suggestions.

Most importantly:

• The tensorflow addons python library is marked as deprecated (https://www.tensorflow.org/addons). Also its not available in conda-forge (which would be a requirement for the Galaxy platform) and I guess conda-forge would not be keen on incorporating deprecated python packages. Would it be possible to replace this?

Other comments:

• would be cool to have the first line #!/usr/bin/env python
• A versioned release via github would be appreciated (e.g. zenodo is fine but it does not allow to download the complete package and also has problems with DOS line endings)
• Create a pepnet module containing utils and the two python scripts (simplifies installation next to other python libraries)
• would be cool to have this pip installable, e.g. by creating setup.py or a project toml file.

Hi, i was trying to follow the tutorial, but i couldn't find the python script "evaluation.py"
python evaluation.py --mgf example.mgf --novorst example_prediction.tsv
Can you tell me where i can download this file?Thanks!

Yafeng