# Using a custom docker image **URL:** \ **Category:** W&B Help\ **Created:** [September 24, 2023, 3:04pm UTC](https://community.wandb.ai/t/using-a-custom-docker-image/5129 "2023-09-24T15:04:02Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![limitingfactor](https://sea2.discourse-cdn.com/flex020/user_avatar/community.wandb.ai/limitingfactor/32/1472_2.png) [@limitingfactor](https://community.wandb.ai/u/limitingfactor)\ **Post date:** [September 24, 2023, 3:04pm UTC](https://community.wandb.ai/t/using-a-custom-docker-image/5129/1 "2023-09-24T15:04:02Z")
Hello, I am getting stuck trying to run a job using a custom docker image and would like some advice. I am trying to use my own custom image based off the nvidia modulus image ([nvcr.io/nvidia/modulus/modulus](http://nvcr.io/nvidia/modulus/modulus)) as I need to newer code than in the base image. This is my dockerfile to generate my image ```auto ARG PYT_VER=23.08 FROM nvcr.io/nvidia/modulus/modulus:$PYT_VER as builder RUN python -m pip install tensorflow RUN python -m pip uninstall -y nvidia-modulus nvidia-modulus.sym nvidia-modulus.launch ENV PYTHONPATH=/modulus/:/modulus-sym/:/modulus-launch/ WORKDIR /modulus-launch/examples/cfd/vortex_shedding_mgn ENTRYPOINT ["sh", "launch.sh"] ``` It just uninstalls the existing modulus code and installs tensorflow. The build isn’t fancy ```auto docker build -t my_modulus:latest -f Dockerfile . ``` For reference my launch.sh ```auto python -m pip uninstall nvidia-modulus nvidia-modulus.sym nvidia-modulus.launch -y cd /modulus/ python -m pip install -e . cd /modulus-sym/ python -m pip install -e . cd /modulus-launch/ python -m pip install -e . cd /modulus-launch/examples/cfd/vortex_shedding_mgn/ git config --global --add safe.directory /modulus-launch pip install wandb --upgrade python /modulus-launch/examples/cfd/vortex_shedding_mgn/wandb_train.py "$@" ``` This makes sure the container has modulus uninstalled and then installs a local version from the mounting points. Finally it run the training script. To test my image I have used ```auto docker run -e WANDB_API_KEY= -e WANDB_DOCKER="my_modulus:latest" --shm-size=1g --ulimit memlock=-1 --ulimit stack=67108864 --runtime nvidia -v :/modulus-launch -v :/modulus -v :/modulus-sym -v :/datasets/ -v :/workspace/ -it --rm my_modulus:latest --project --entity ``` This works fine and I get a job created on wandb. I have setup a docker queue ```auto env: - PYTHONPATH=/modulus/:/modulus-sym/:/modulus-launch/ gpus: all volume: - :/modulus-launch - :/modulus - :/modulus-sym - :/datasets/ - :/workspace/ builder: accelerator: base_image: my_modulus:latest ``` I have tried with and without the builder. When launching the job from the website I use these options ```auto { "args": [ "--project", "", "--entity", "" ], "run_config": { "epochs": 25, "ckpt_path": "/workspace/checkpoints_training_6" }, "entry_point": [ ] } ``` This is to test changing the number of epoch and to save the checkpoints to a different folder. The error I get is this ```auto wandb: launch: Launching run in docker with command: docker run --rm -e WANDB_BASE_URL=https://api.wandb.ai -e WANDB_API_KEY -e WANDB_PROJECT= -e WANDB_ENTITY= -e WANDB_LAUNCH=True -e WANDB_RUN_ID=7dyx6mlk -e WANDB_USERNAME= -e WANDB_CONFIG='{"epochs": 25, "ckpt_path": "/workspace/checkpoints_training_6"}' -e WANDB_ARTIFACTS='{"_wandb_job": "//job-_wandb_train.py:latest"}' --env PYTHONPATH=/modulus/:/modulus-sym/:/modulus-launch/ --gpus all --volume :/modulus-launch --volume :/modulus --volume :/modulus-sym --volume :/datasets/ --volume :/workspace/ <>_wandb_train.py:7864ffe4 Traceback (most recent call last): File "<>/examples/cfd/vortex_shedding_mgn/wandb_train.py", line 20, in import torch ModuleNotFoundError: No module named 'torch' ``` Looking at the documentation it isn’t clear to me how to do this correctly. If anyone has some advice on how to use a docker image similar to what I have created above I would be grateful. Thanks LimitingFactor ---
**Author:** ![uma-wandb](https://avatars.discourse-cdn.com/v4/letter/u/b77776/32.png) [@uma-wandb](https://community.wandb.ai/u/uma-wandb)\ **Post date:** [September 28, 2023, 6:39pm UTC](https://community.wandb.ai/t/using-a-custom-docker-image/5129/2 "2023-09-28T18:39:22Z")
Hey @limitingfactor - in your setup are you explicitly installing pytorch anywhere? We don’t automatically do this, so I just wanted to check on what happens if you explicitly download this beforehand or include it in your setup. ---
**Author:** ![limitingfactor](https://sea2.discourse-cdn.com/flex020/user_avatar/community.wandb.ai/limitingfactor/32/1472_2.png) [@limitingfactor](https://community.wandb.ai/u/limitingfactor)\ **Post date:** [September 29, 2023, 10:52pm UTC](https://community.wandb.ai/t/using-a-custom-docker-image/5129/3 "2023-09-29T22:52:23Z")
Hi @uma-wandb, I am using the Nvidia modulus image as a starting point which already has pytorch installed in it. ---
**Author:** ![uma-wandb](https://avatars.discourse-cdn.com/v4/letter/u/b77776/32.png) [@uma-wandb](https://community.wandb.ai/u/uma-wandb)\ **Post date:** [November 14, 2023, 6:13am UTC](https://community.wandb.ai/t/using-a-custom-docker-image/5129/4 "2023-11-14T06:13:55Z")
hey @limitingfactor - were you able to get this up and running? have you run into any further issues? ---
**Author:** ![system](https://us1.discourse-cdn.com/flex020/uploads/wandb/original/1X/366b649231631dbab896843020da0056074ac79d.png) [@system](https://community.wandb.ai/u/system)\ **Post date:** [January 13, 2024, 6:14am UTC](https://community.wandb.ai/t/using-a-custom-docker-image/5129/5 "2024-01-13T06:14:16Z")
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.