Dockerfile issues pulling assets from S3

[ZaheedaT]

=> ERROR [12/12] RUN dvc pull 1.7s

[12/12] RUN dvc pull:

17 1.298 WARNING: Some of the cache files do not exist neither locally nor on remote. Missing cache files:

[GokuMohandas]

Hi @ZaheedaT, the lessons are meant to be taken in sequential order and we create many of the artifacts needed in previous lessons so that's why this error is happening. We create the necessary data assets and version them with DVC in a prior lesson so this Docker lesson won't run properly if you don't follow the previous lessons.

[ZaheedaT]

Hi there, I ran all the lessons, out of Docker as well, and still get this error. I ran this whole workflow first

python tagifai/main.py load-data
python tagifai/main.py label-data --args-fp="config/args.json"
python tagifai/main.py optimize --args-fp="config/args.json" --study-name="optimization" --num-trials=10
python tagifai/main.py train-model --args-fp="config/args.json" --experiment-name="baselines" --run-name="sgd"
python tagifai/main.py predict-tag --text="Transfer learning with transformers for text classification."
``

[GokuMohandas]

Unfortunately those commands only train a model ready for inference and the DVC steps required are not covered in those commands above. The versioning lesson will have the steps required to version and store the data in order for Docker to be able to pull them.

[ZaheedaT]

Do mean using the exact version of Docker? Or Python version?

There's an issue with the Dockerfile, there is no directory called "stores" in the repo

[ZaheedaT]

I followed all the steps from https://madewithml.com/courses/mlops/versioning/

Still getting the same error. I'll leave the course here. Thanks

[GokuMohandas]

Hi @ZaheedaT , unfortunately MLOps concepts are quite interweaved and cannot be run in isolation. For example, the stores directory is created in a different lesson (you can see it’s created when config/config.py is run, which we do in one of the early lessons in scripting). You absolutely need to run the course in sequential order for things to work and cannot skip to one of the later lessons. The course has been successfully completed by tens of thousands over the past year and all were able to successfully complete it across multiple OS and we have thorough tests to ensure everything always runs with all updates to the course. I’m happy to help if you run into specific issues but the course absolutely needs to be taken sequentially.