agileventures / automatedpremiummodel Goto Github PK
View Code? Open in Web Editor NEWAutomation of the AV prediction algorithm for premium signup
License: Other
Automation of the AV prediction algorithm for premium signup
License: Other
Ping #data-mining with the results of the model but no emails.
Ping @tansaku with results of the model including emails
Make README beautiful for github view with better markdown use.
Add instructions for Azure VM related issues like getting the model working on azure vm as well as adding other users to the vm who can sudo.
using dokku and friends
I've been trying to ensure that someone who is listed in the premium API as premium is excluded from the weekly suggestions, but putting their AV email as 'email' and their slack email as 'alias_email' does not appear to be working
file is /var/lib/dokku/data/storage/email_aliases.csv with headers "email,alias_email,,,,,"
https://github.com/AgileVentures/AutomatedPremiumModel/blob/master/basic_functionality_so_far.R#L103
this line isnt' needed and the variable probs, which isn't defined elsewhere, makes it trip over itself
we can probably consolidate the test_token and the api_token into just one bot token with requisite permissions.
We seem to be getting newer premium members in the predictions after we dockerized. Possibly related to the change in the data to be in a folder data and not updating the update_premimum_members script.
quick fix for production
some mentors (e.g. @joaopapereira ) have suggested that along with the suggestions that it might be useful to have information about when a user joined slack for the purpose of informing the way in which the mentor reaches out to the member
Right now, we need to document and support in the repo how we are installing a crontab task to automatically run every sunday morning and ping slack with the results.
The recently introduced forecast bot showed up in the predictions for some reason, even though we are trying to filter out bots.
might be nice
Use the newly available api on WSO for premium signup data to update the data weekly.
*Distinct from updating the legacy data
currently, the functions surrounding this seem to work reasonably well but we need to test this spikey behavior, as well as maybe improve it.
Currently we have a manual process to assign mentors to reach out to premium upgrade prospects. We use the datamining data post from the bot and spreadsheet from history in order to work out which mentors to assign to which prospects:
post to #datamining data
==> a,b,c,d,e,f,g,h
=====================
mentors
previously predicted - 2 months?:
previously predicted - 2 months?:
new
good to flag binghamton students - can usually tell from emails
to make their pm linkable
...
there are a lot of them who are rather active at particular times of year, and they seem uniformly disinterested in AgileVentures, beyond what's required for their course - could be helpful to just exclude anyone whose address ends in binghamton.edu
For instance, determine the previous saturday and go back three weeks from there. This is partially implemented in a function finding previous saturday already
get current premiums from api
could dump to csv
could connect up directly?
...
this might be better long term, if for instance, rather than passing in a token at each turn, we just pass it during initialization of "object"
perhaps the current model should be put into its own module or file at least?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.