Dear organizers, our team GeckoAD (ID 9760701) submitted a docker container, but couldn’t generate the prediction.csv file. We have run the container on another server with success. We reviewed the log, and it seems to be saying that it uses 1 core, 100MB of ram, and tmpsize of 1024. For our model we use 25 threads, 100 GB of temp dir size, 192 GB of RAM. Is it possible to modify the container resources for our case? Or if there is a different error for the submitted entry, could you send us more information about our error?
Thank you very much.
Created by Allan Miyazono tach Thank you very much. Hi @tach ,
The log file you viewed is most likely for the workflow, rather than for the Docker container. (Due to the current sensitivity of the data, container logs are not made available to participants. However, we can share snippets of the logs to help with debugging.) For your subID 9760701, the following error was received:
```
Traceback (most recent call last):
File "/app/main.py", line 173, in
sys.exit(main())
^^^^^^
File "/app/main.py", line 44, in main
preprocess_metadata_mtg = preprocess_metadata(f"{data_dir}/metadata_pre.csv", f"{data_dir}/metadata.csv", csv=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/python/gecko/src/gecko/prepro_metadata.py", line 47, in preprocess_metadata
modified_df['Cognitive_Status_encoded'] = label_encoder.fit_transform(modified_df['Cognitive Status'])
~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^
File "/app/python/gecko/.venv/lib/python3.12/site-packages/pandas/core/frame.py", line 4107, in __getitem__
indexer = self.columns.get_loc(key)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/app/python/gecko/.venv/lib/python3.12/site-packages/pandas/core/indexes/base.py", line 3819, in get_loc
raise KeyError(key) from err
KeyError: 'Cognitive Status'
```
From [another thread](https://www.synapse.org/Synapse:syn66496696/discussion/threadId=12437&replyId=35782?replyid=35782) regarding "Cognitive Status":
> Cognitive Status is only available in the training dataset due to its correlation with the metrics that are being predicted. All other variables/data (except those that are being predicted) are available.
Hope this helps!