search

sfbrigade / datasci-open-payments

Repository for UCSF Mental Health Hackathon 2019 Exploring Considerations in Finance and Prescriptions for Mental Health Specialists
1 stars 0 forks source link

Create CLI Tool #11

Open VincentLa14 opened 5 years ago

VincentLa14 commented 5 years ago

https://click.palletsprojects.com/en/4.x/setuptools/

benjaminbang987 commented 5 years ago

Made a cli here: https://github.com/sfbrigade/c4sfcensusgeocoder Nvm, need to update the cli to take in the csv file or dataframe.

benjaminbang987 commented 5 years ago

@VincentLa - given that the censusgeocoder is already doing the csv file input parser, perhaps we should aim to do something different? Or perhaps we could just move on with the dataframe input?

VincentLa commented 5 years ago

I think there are several things we can do even if the censusgeocoder already does CSV input. Like maybe we can do better error handling, better treatment of null values. Also how does censusgeocoder work in terms of returning fields? Does it just return all the fields and then the user has to parse JSON? Maybe we can add an option where user can specify specifically the fields they want and our tool returns those fields as separate columns

On Wed, May 8, 2019, 7:29 PM benjaminbang987 notifications@github.com wrote:

@VincentLa https://github.com/VincentLa - given that the censusgeocoder is already doing the csv file input parser, perhaps we should aim to do something different? Or perhaps we could just move on with the dataframe input?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/sfbrigade/datasci-open-payments/issues/11#issuecomment-490719211, or mute the thread https://github.com/notifications/unsubscribe-auth/AC6S7R4GUBNWY4ENPLLSYYLPUOEAHANCNFSM4HIJTRZQ .

benjaminbang987 commented 5 years ago

I guess what I'm suggesting is that there are two paths going forward

  1. Besides the cli, the package still works. Now that we have the geocodes in the dataframe, we move onto the next steps of actually using that information to pull in demographic information as features and try to actually make some insights.
  2. Spend a bit more time closing the loop on the cli side of the package and then move on to the analysis.

I was just trying to weigh the options and see where we should be focusing next.