Docker image for multiple evalulation broken

[Extirpater]

After generating a file for java: generations_multiple-java.json
and building a docker image for multiple eval: sudo make DOCKERFILE=Dockerfile-multiple all

I get the following error when running the dummy model test as well as my own evaluation for my generated java json file.

`ubuntu@192-222-52-130:~/bigcode-evaluation-harness$ sudo make DOCKERFILE=Dockerfile-multiple all
docker build -f Dockerfile-multiple -t evaluation-harness-multiple .
[+] Building 82.6s (8/9) docker:default
[+] Building 82.8s (8/9) docker:default
[+] Building 112.8s (10/10) FINISHED docker:default
=> [internal] load .dockerignore 0.0s
=> => transferring context: 2B 0.0s
=> [internal] load build definition from Dockerfile-multiple 0.0s
=> => transferring dockerfile: 327B 0.0s
=> [internal] load metadata for ghcr.io/nuprl/multipl-e-evaluation@sha256:11864ca95774df16c34b4cd1eac23 0.0s
=> [internal] load build context 0.7s
=> => transferring context: 3.49MB 0.7s
=> [1/5] FROM ghcr.io/nuprl/multipl-e-evaluation@sha256:11864ca95774df16c34b4cd1eac231f9e5466c7ea38dac9 0.0s
=> CACHED [2/5] COPY . /app 0.0s
=> CACHED [3/5] WORKDIR /app 0.0s
=> CACHED [4/5] RUN test -f /app/generations.json && rm /app/generations.json || true 0.0s
=> [5/5] RUN pip3 install . 85.0s
=> exporting to image 26.9s
=> => exporting layers 26.9s
=> => writing image sha256:3408cb4ed07bb80070a2fa6b229921776301f1e7ace128968297dcc9bf950e87 0.0s
=> => naming to docker.io/library/evaluation-harness-multiple 0.0s
docker run -v /home/ubuntu/bigcode-evaluation-harness/tests/docker_test/test_generations.json:/app/test_generations.json:ro \
-it evaluation-harness-multiple python3 main.py --model dummy_model --tasks humaneval --limit 4
--load_generations_path /app/test_generations.json --allow_code_execution
.....
[--save_generations_path SAVE_GENERATIONS_PATH] [--save_references]
[--save_references_path SAVE_REFERENCES_PATH] [--prompt PROMPT]
[--max_memory_per_gpu MAX_MEMORY_PER_GPU] [--check_references]
main.py: error: unrecognized arguments: python3 main.py
make: *** [makefile:14: test] Error 2

`

Is there something wrong with the docker image?

[Vermouth7]

I had the same problem. I think it's that the default execution command for this image is python3 main.py, so for running in docker, you just need pass the parameters. So the command to run should be like:

docker run -v /home/ubuntu/bigcode-evaluation-harness/tests/docker_test/test_generations.json:/app/test_generations.json:ro \
-it evaluation-harness-multiple --model dummy_model --tasks humaneval --limit 4
--load_generations_path /app/test_generations.json --allow_code_execution