-
Notifications
You must be signed in to change notification settings - Fork 1
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Failing test in new installation #14
Comments
I've discovered a workaround for the crash. My .pdb files are named with a unique convention. Renaming the files to a pattern matching the RCSB convention (e.g. 1XXX.pdb) allows DeepRank-GNN-esm to proceed. |
Thank you for your interest in our tool! |
For the record, the first issue I reported here remains unsolved (the failing test). |
Hi there, In what part is the code failing? The assert check ensures embedding is added correctly to the graph. If you re-name your pdb file as xxxx.pdb, the code should run fine. Also, try to use pdb_tidy command from pdb-tools to clean the pdb file if it is not the standard format. Best, |
This is the FAILED tests/test_biowrappers.py::TestBioWrappers::test_depth_res - Failed: Unexpected success |
Here's the complete output of a test failure from a fresh installation on a new Ubuntu machine with a 4090. The main failure seems to be in TestBioWrappers.test_depth_res. Let me know if there are any other logs I can provide. |
Thank you for publishing such a cool tool! I'm running into some issues installing and using the scoring function.
First, I'm seeing tests failing in a fresh installation:
FAILED tests/test_biowrappers.py::TestBioWrappers::test_depth_res - Failed: Unexpected success
=================================================== 1 failed, 21 passed, 3 xfailed, 144223 warnings in 347.77s (0:05:47) ================================================
DeepRank-GNN-esm_test.txt
When I attempt to use the model, an assertion error is thrown suggesting the embeddings are invalid:
$ deeprank-gnn-esm-predict ./053_5.pdb C A ../paper_pretrained_models/scoring_of_docking_models/gnn_esm/treg_yfnat_b64_e20_lr0.001_foldall_esm.pth.tar 2024-12-03 20:51:26,609 predict:60 INFO - Setting up workspace - /share/data/public/edlunde/git/DeepRank-GNN-esm/test/053_5-gnn_esm_pred_C_A 2024-12-03 20:51:26,616 predict:69 INFO - Renumbering PDB file. 2024-12-03 20:51:26,685 predict:112 INFO - Reading sequence of PDB 053_5.pdb 2024-12-03 20:51:26,716 predict:138 INFO - Generating embedding for protein sequence. 2024-12-03 20:51:26,716 predict:139 INFO - ################################################################################ 2024-12-03 20:51:33,577 predict:154 INFO - Read /share/data/public/edlunde/git/DeepRank-GNN-esm/test/053_5-gnn_esm_pred_C_A/all.fasta with 2 sequences 2024-12-03 20:51:33,586 predict:164 INFO - Processing 1 of 1 batches (2 sequences) 2024-12-03 20:51:43,586 predict:206 INFO - ################################################################################ 2024-12-03 20:51:43,669 predict:211 INFO - Generating graph, using 31 processors Graphs added to the HDF5 file Traceback (most recent call last): File "~/conda/envs/deeprank-gnn-esm-cpu-env/bin/deeprank-gnn-esm-predict", line 8, in <module> sys.exit(main()) File "~/conda/envs/deeprank-gnn-esm-cpu-env/lib/python3.9/site-packages/deeprank_gnn/predict.py", line 335, in main graph = create_graph(pdb_path=pdb_file.parent, workspace_path=workspace_path) File "~/conda/envs/deeprank-gnn-esm-cpu-env/lib/python3.9/site-packages/deeprank_gnn/predict.py", line 216, in create_graph GraphHDF5( File "~/conda/envs/deeprank-gnn-esm-cpu-env/lib/python3.9/site-packages/deeprank_gnn/GraphGenMP.py", line 141, in __init__ self._add_embedding(outfile=outfile, pdbs=pdbs, embedding_path=embedding_path) File "~/conda/envs/deeprank-gnn-esm-cpu-env/lib/python3.9/site-packages/deeprank_gnn/GraphGenMP.py", line 247, in _add_embedding assert not torch.all(torch.eq(embedding_tersor, 0)) AssertionError
I am running in CPU mode, for what it's worth, but the test failure seems unrelated.
The text was updated successfully, but these errors were encountered: