angeloschert

There may be some components that are hard to develop using pure Django, so we could consider using embedded H5/JS in this case.

Our entire team is using this IDE for almost all time at university or during internships, it supports multiple useful plug-ins and supports remote ssh access for dev in a remote server.

We will use PyTorch to implement the neural network model for the prediction of product subscription rates given the prospective clients. Numpy is well compatible with PyTorch in the array processing, and also we need to do numerical analysis with the output and the raw data so Numpy is crucial

Slack is more work-based and seems more professional when it comes to video calls, code sharings and message threading. Besides, I could add Slack workflow/bots to our app during the testing phase and report any errors msgs in our work channels.

Our app will be in its initial stage while it is nearly completed and will not involve massive data IO or storage. Heroku provides a free service-based deployment platform for our web-based app and it has good compatibility with Django.

We use them for version control, issue tracking, and milestone monitoring. Our dev process involves branching dev and code review and Git + Github are the most prevalent tools for our work.

We will need to test the validity of RESTful APIs during our development. There will be a large amount of data requests during the testing phase. Postman provides a clear UI to edit POST messages and view the logs.

We use it for messaging queue since we need a pipeline to process incoming product data. Besides, Spark has built-in analysis tool which is helpful.

We are focusing on data processing with a single product and could easily be drafted with a fixed data scheme. It will be much nicer to use MySQL over any NoSQL.

We use Python as the primary language and Django as the framework for both UI and backend business logics, since Django is more easily deployed on Heroku and we have previous experience with it.