initialdlab / aipro Goto Github PK

Some outdated modules from requirements.txt are either not used or unsafe. These need to be removed.

Once models are making predictions, it's important for the user to be able to give feedback about whether or not the prediction was correct. This way, data scientists can get labeled data as feedback for retraining their models.

Currently, the pipeline functionality supports transformations and predictions with different models. We need to be able to create training pipelines as well.

Thanks.

I am proposing that we make the file structure of the code as follows:

aipro

• |- docs/ Wiki and other documentation
• |- web/ Webserver and UI components
• |- server/ or src/ AIPro backend code resides here
  -- |- model.py
  -- |- main.py
  -- |- api.py
  -- |- data_sources/
  -- |- ........
  -- |- ........
• |- install.sh Installation and running files.
• |- README.md
• |- requirements.txt
• |- docker-compose.yml

That will encapsulate and make separation of concern lucid.

Integrate with Travis for CI/CD pipeline

Currently, the code for tree representation in the UI is way too complicated and prone to bugs. Simplifying the tree would make this a lot easier to maintain.

Need unit tests and a testing process for front-end components for CI/CD

We really should have some unit tests for the backend components and a testing process.

Should be updated to at least Python 3.6