🐛 Bug: Fetching model eos54ij got "Not a directory" error

[RoseZhen2022]

Describe the bug.

I have fetch model eos54ij multiple times and tried to fetch it with or without export log, both got same error: NotADirectoryError: [Errno 20] Not a directory

Describe the steps to reproduce the behavior

1. ersilia -v fetch eos54ij | tee -a eos54ij.log 2>&1
2. ersilia -v fetch eos54ij

Expected behavior.

Hope to fetch the model and do next step

Screenshots.

Operating environment

macOS Monterey

Additional context

The log file created is nothing inside, so I didn't put the log file.

[GemmaTuron]

Hello @RoseZhen2022

I just run it on my linux machine without problems:

17:30:30 | DEBUG | [{'input': {'key': 'LUHMMHZLDLBAKX-UHFFFAOYSA-N', 'input': 'CC1C2C(CC3(C=CC(=O)C(=C3C2OC1=O)C)C)O', 'text': 'CC1C2C(CC3(C=CC(=O)C(=C3C2OC1=O)C)C)O'}, 'output': {'outcomes': [None, None, 0.8, 0.0, 0.2, 0.26]}}, {'input': {'key': 'QRXWMOHMRWLFEY-UHFFFAOYSA-N', 'input': 'C1=CN=CC=C1C(=O)NN', 'text': 'C1=CN=CC=C1C(=O)NN'}, 'output': {'outcomes': [None, None, 0.9, 0.0, 0.25, 0.26]}}] 17:30:30 | DEBUG | Latest meta: {'outcomes': None} 17:30:30 | DEBUG | outcomes : {'type': 'mixed_array', 'shape': (6,)} 17:30:30 | DEBUG | Meta k: None 17:30:30 | DEBUG | Schema: {'input': {'key': {'type': 'string'}, 'input': {'type': 'string'}, 'text': {'type': 'string'}}, 'output': {'outcomes': {'type': 'mixed_array', 'shape': (6,), 'meta': None}}} 17:30:30 | DEBUG | {'input': {'key': {'type': 'string'}, 'input': {'type': 'string'}, 'text': {'type': 'string'}}, 'output': {'outcomes': {'type': 'mixed_array', 'shape': (6,), 'meta': None}}} 17:30:30 | DEBUG | API schema saved at /home/gturon/eos/dest/eos54ij/api_schema.json 17:30:30 | INFO | Fetching eos54ij done successfully: 0:02:53.585336 👍 Model eos54ij fetched successfully!

Can you let me know which:

• Conda version
• Python version you are using?

[GemmaTuron]

Also, the "Not a Directory" error seems like you are running things from a location that ersilia is unable to find. Can you try to run the command from your desktop?

[GemmaTuron]

Hi @RoseZhen2022 Have you been able to check my suggestions?

[RoseZhen2022]

Hi @GemmaTuron Sorry for the late reply. As the operating system I used before has many problems. So recently I changed to another system Ubuntu 22.04.1 LTS and set the new Erisilia working environment. Using this new system, I have successfully predicted 3 of 4 models you assigned me before and I have already added the output csv files to drive. However, only this model eos54ij was fetched failed with "EmptyOutputError". And here is the log of this model. eos54ij.log

[RoseZhen2022]

Hi @GemmaTuron , to make this more clear, I create a new issue https://github.com/ersilia-os/ersilia/issues/386. I will close this issue.