MannLabs / alphapeptdeep

Deep learning framework for proteomics
Apache License 2.0
102 stars 20 forks source link

Value error + Peptdeep is trying to use multiprocess even when it is not selected #123

Closed Maithy15 closed 7 months ago

Maithy15 commented 7 months ago

Hi,

Thanks for the latest update.

I am trying it but it seems to get stuck at one point with no output. I have the log pasted here. Additionally I have uploaded the settings file. Please et me know if there is any mistake in the setting.

peptdeep_library_2023-12-29--10-08-25.862954.yaml.zip

Starting PeptDeep Web Server on port 10077 ...

You can now view your Streamlit app in your browser.

Local URL: http://localhost:10077 Network URL: http://193.174.62.4:10077

[PeptDeep] Starting a new job 'C:\Users\m284y/peptdeep/tasks/queue\peptdeep_library_2023-10-05--19-40-40.124913.yaml'...[PeptDeep] Predicting library ... 2023-12-29 10:07:32> [PeptDeep] Running library task ... 2023-12-29 10:07:32> Input files (fasta): ['E:/FASTA/HUMAN_070321/7March21_uniprot-proteome_UP000005640+reviewed_yes.fasta'] 2023-12-29 10:07:32> Platform information: 2023-12-29 10:07:32> system - Windows 2023-12-29 10:07:32> release - 10 2023-12-29 10:07:32> version - 10.0.19045 2023-12-29 10:07:32> machine - AMD64 2023-12-29 10:07:32> processor - AMD64 Family 23 Model 49 Stepping 0, AuthenticAMD 2023-12-29 10:07:32> cpu count - 128 2023-12-29 10:07:32> ram - 234.3/255.9 Gb (available/total) 2023-12-29 10:07:32> 2023-12-29 10:07:32> Python information: 2023-12-29 10:07:32> alphabase - 1.2.0 2023-12-29 10:07:32> alpharaw - 0.4.0 2023-12-29 10:07:32> biopython - 2023-12-29 10:07:32> click - 8.1.7 2023-12-29 10:07:32> lxml - 4.9.4 2023-12-29 10:07:32> numba - 0.58.1 2023-12-29 10:07:32> numpy - 1.26.2 2023-12-29 10:07:32> pandas - 2.1.4 2023-12-29 10:07:32> peptdeep - 1.1.1 2023-12-29 10:07:32> psutil - 5.9.7 2023-12-29 10:07:32> pyteomics - 4.6.3 2023-12-29 10:07:32> python - 3.9.18 2023-12-29 10:07:32> scikit-learn - 1.3.2 2023-12-29 10:07:32> streamlit - 1.29.0 2023-12-29 10:07:32> streamlit-aggrid - 2023-12-29 10:07:32> torch - 2.1.2 2023-12-29 10:07:32> tqdm - 4.66.1 2023-12-29 10:07:32> transformers - 4.36.2 2023-12-29 10:07:32> 2023-12-29 10:07:34> xxx/library.tsv does not exist, use default IRT_PEPTIDE_DF to translate irt 2023-12-29 10:07:34> Generating the spectral library ... 2023-12-29 10:11:37> Loaded 23643549 precursors. 2023-12-29 10:13:50> Predicting RT/IM/MS2 for 19882736 precursors ... 2023-12-29 10:13:50> Using multiprocessing with 100 processes ... 2023-12-29 10:13:50> Predicting rt,mobility,ms2 ... Exception in thread Thread-1: Traceback (most recent call last): File "threading.py", line 980, in _bootstrap_inner File "threading.py", line 917, in run File "multiprocessing\pool.py", line 519, in _handle_workers File "multiprocessing\pool.py", line 499, in _wait_for_updates File "multiprocessing\connection.py", line 879, in wait File "multiprocessing\connection.py", line 811, in _exhaustive_wait ValueError: need at most 63 handles, got a sequence of length 102 47%|████████████████████████████████████████████████████████▍ | 101/213 [1:11:24<1:31:48, 49.18s/it]

Maithy15 commented 7 months ago

Just give you an update, I tried it one another computer and it working fine

2023-12-29 12:01:14> Downloading pretrained_models.zip ... 2023-12-29 12:01:17> The pretrained models had been downloaded in C:\Users\m284y/peptdeep\pretrained_models\pretrained_models.zip Starting PeptDeep Web Server on port 10077 ...

You can now view your Streamlit app in your browser.

Local URL: http://localhost:10077 Network URL: http://172.22.62.33:10077


[PeptDeep] Waiting for tasks ...


[PeptDeep] Starting a new job 'C:\Users\m284y/peptdeep/tasks/queue\peptdeep_library_2023-12-29--12-05-00.476766.yaml'...[PeptDeep] Predicting library ... 2023-12-29 12:05:02> [PeptDeep] Running library task ... 2023-12-29 12:05:02> Input files (fasta): ['D:/FASTA/Fasta_2021_Copy/HUMAN_070321/7March21_uniprot-proteomeUP000005640+reviewed_yes.fasta'] 2023-12-29 12:05:02> Platform information: 2023-12-29 12:05:02> system - Windows 2023-12-29 12:05:02> release - 10 2023-12-29 12:05:02> version - 10.0.19045 2023-12-29 12:05:02> machine - AMD64 2023-12-29 12:05:02> processor - AMD64 Family 25 Model 8 Stepping 2, AuthenticAMD 2023-12-29 12:05:02> cpu count - 64 2023-12-29 12:05:02> ram - 119.2/127.8 Gb (available/total) 2023-12-29 12:05:02> 2023-12-29 12:05:02> Python information: 2023-12-29 12:05:02> alphabase - 1.2.0 2023-12-29 12:05:02> alpharaw - 0.4.0 2023-12-29 12:05:02> biopython - 2023-12-29 12:05:02> click - 8.1.7 2023-12-29 12:05:02> lxml - 4.9.4 2023-12-29 12:05:02> numba - 0.58.1 2023-12-29 12:05:02> numpy - 1.26.2 2023-12-29 12:05:02> pandas - 2.1.4 2023-12-29 12:05:02> peptdeep - 1.1.1 2023-12-29 12:05:02> psutil - 5.9.7 2023-12-29 12:05:02> pyteomics - 4.6.3 2023-12-29 12:05:02> python - 3.9.18 2023-12-29 12:05:02> scikit-learn - 1.3.2 2023-12-29 12:05:02> streamlit - 1.29.0 2023-12-29 12:05:02> streamlit-aggrid - 2023-12-29 12:05:02> torch - 2.1.2 2023-12-29 12:05:02> tqdm - 4.66.1 2023-12-29 12:05:02> transformers - 4.36.2 2023-12-29 12:05:02> 2023-12-29 12:05:04> xxx/library.tsv does not exist, use default IRT_PEPTIDE_DF to translate irt 2023-12-29 12:05:04> Generating the spectral library ... 2023-12-29 12:11:00> Loaded 41331786 precursors. 2023-12-29 12:14:02> Predicting RT/IM/MS2 for 38231621 precursors ... 2023-12-29 12:14:03> Predicting RT ... 28%|████████████████████████████▋ | 8/29 [20:08<1:03:23, 181.12s/it]

jalew188 commented 7 months ago

Hi @Maithy15, Python only allows maximal number of 64 threads:

ValueError: need at most 63 handles, got a sequence of length 102