I received a ModuleNotFoundError for 'aicrowd_gym' for the demixing validation stage. The repo2docker image builds perfectly fine and runs the 'evaluate_locally.py' to demix and score my model. I wanted to know if this needs to be included in the requirements.txt (didn’t see it mentioned anywhere). Or is this a bug on the backend?
Additionally, are there any other packages that need to be included in the requirement.txt that are independent of my code to get the "/home/aicrowd/client_launcher.py" and other aicrowd files to function properly?
########################################
# rollout-music-demixing-validation #
########################################
Traceback (most recent call last):
File "/home/aicrowd/client_launcher.py", line 3, in <module>
from aicrowd_gym.clients.zmq_oracle_client import ZmqOracleClient
ModuleNotFoundError: No module named 'aicrowd_gym'
UPDATE
I tried submitting again and included the 'aicrowd_gym' package in my requirements and came up with nearly the same result. I would like to avoid wasting any more submissions, so I will have to wait until this can be addressed.
########################################
# rollout-music-demixing-validation #
########################################
Traceback (most recent call last):
Traceback (most recent call last):
File "/home/aicrowd/client_launcher.py", line 3, in <module>
File "/home/aicrowd/client_launcher.py", line 3, in <module>
from aicrowd_gym.clients.zmq_oracle_client import ZmqOracleClient
from aicrowd_gym.clients.zmq_oracle_client import ZmqOracleClient
ModuleNotFoundError: No module named 'aicrowd_gym.clients'
ModuleNotFoundError: No module named 'aicrowd_gym.clients'
Thanks @kimberley_jensen! I’ll give it a try (if I have any submissions left) and see if it works. Otherwise, hopefully someone from AIcrowd can take a look on Monday.
Hi @crlandsc , our evaluation setup doesn’t work with environment.yml directly added to the repo. If you need to use environment.yml, you need to add a Dockerfile. If possible, avoid environment.yml altogether and directly install cudatoolkit in the Dockerfile/select a base Docker image with the preferred cudatoolkit.
@dipam Forgive me for the confusion, but the instructions on the webpage are counter to what you are describing here (it even notes that environment.yml is the recommended method):
"SUBMISSION ENVIRONMENT CONFIGURATION You can specify your software environment by using Dockerfile, environment.yml, requirements.txt and so on. The list of available methods are available here."
So just to be totally clear, do you HAVE to use the Dockerfile? The options also say that you can just use requirements.txt. Why is there an environment.yml included in the starter kit?
Apologies for the confusion. We are making some changes to our evaluation setup, hence the difference from the documentation provided. However, I understand that since the challenge is ending using environment.yml might be important, hence we’ve added back the support for it.
I’ve made some minor changes to your environment.yml and made a submission. I believe now there is some error for the code to run, please check the logs.
@dipam no problem, thanks for updating and resubmitting for me. I ended up creating a dockerfile from the one you shared anyway to test things on my end. Good to know both methods work now.