search

christopherjenness / NBA-prediction

Predict scores of NBA games using regularized matrix completion
152 stars 43 forks source link

Can't run model without predictions.csv #3

Closed mc-buckets closed 7 years ago

mc-buckets commented 7 years ago

Hi! I'm trying to use your model - very cool work - but can't initialize the object without a predictions.csv file. Is that something I need to manually pull before running?

christopherjenness commented 7 years ago

Hi, this is a common problem people are reporting. Preidctions.csv is written when you initialize the model with update=True. I put that in the sample code in the readme, but I should probably change the sample code at the bottom of the script so people understand. Thanks!

vladiki commented 7 years ago

In model.py you should instantiate the class with update=True. It'll pull down the latest data and generate the right files. You may also need to install a couple packages for R to get it to run correctly

christopherjenness commented 7 years ago

Oh, thanks vladiki, I forgot I used some R packages. I'll add that to the readme.

mc-buckets commented 7 years ago

A requirements.txt file would be great, too!

christopherjenness commented 7 years ago

Yeah, I just pushed the requirements.txt. Not familiar with the idiomatic way to specify which R packages are used though..