Closed chtsai0105 closed 2 years ago
Hello @chtsai0105, I believe this is an issue we've previously encountered.
Issue: https://github.com/KwanLab/Autometa/issues/285 -> PR: https://github.com/KwanLab/Autometa/pull/286
Good to know that downgrading joblib
also fixes the issue.
Thank you Evan! Since there is already a fix, I'm closing this issue.
Current Behavior
Steps to Reproduce
An error occurred immediately while running autometa-binning step when installed from conda/mamba.
Expected Behavior
Expected to see the help page.
Environment Information
autometa-config --print
Not sure what happening... ```bash [09/17/2022 11:42:50 PM INFO] autometa.config.utilities: Updating default.config ([common],home_dir): /home/chtsai/miniconda3/envs/autometa/lib/python3.9/site-packages [09/17/2022 11:42:50 PM DEBUG] autometa.config.utilities: updated /home/chtsai/miniconda3/envs/autometa/lib/python3.9/site-packages/autometa/config/default.config [common] option: home_dir : /home/chtsai/miniconda3/envs/autometa/lib/python3.9/site-packages Traceback (most recent call last): File "/home/chtsai/miniconda3/envs/autometa/bin/autometa-config", line 10, in
sys.exit(main())
File "/home/chtsai/miniconda3/envs/autometa/lib/python3.9/site-packages/autometa/config/utilities.py", line 237, in main
cfg, environ_satisfied = environ.configure(cfg)
File "/home/chtsai/miniconda3/envs/autometa/lib/python3.9/site-packages/autometa/config/environ.py", line 265, in configure
versions = get_versions()
File "/home/chtsai/miniconda3/envs/autometa/lib/python3.9/site-packages/autometa/config/environ.py", line 229, in get_versions
version = get_version()
File "/home/chtsai/miniconda3/envs/autometa/lib/python3.9/site-packages/autometa/config/environ.py", line 148, in bowtie2
return stdout.decode().split()[2]
IndexError: list index out of range
```
Possible solution
Actually I've run 2.1.0 version successfully before. I compared the previous build and current build and found that the package
joblib
have been updated recently. (1.1.0 -> 1.2.0) I downgraded thejoblib
to 1.1.0 and everything is good now. Maybe have to look into what was changed in the newerjoblib
update or simply pin thejoblib
to 1.1.0.