mthorrell

Interesting, it does work for me now. In any case, this is on the GLM Github homepage (https://github.com/JuliaStats/GLM.jl). Perhaps that should be edited? ``` julia> newX = DataFrame(X=[2,3,4]); julia> predict(OLS,...

@baozzhao , if you're still interested, there's a project I'm working on that combines pytorch and xgboost ([here](https://github.com/mthorrell/gboost_module)). This makes ordinal regression fairly straightforward to fit using xgboost as you...

UPDATE -- going to go a general integration route, so we can support Poisson processes as well ```python from __future__ import annotations import warnings from typing import List, Dict, Any...

> Pytorch loss functions accept weights so this may be the solution to this problem. Yes, I think this would be the "right" way to do this. Weights, as I...

Just trying to read the code there, it looks like potentially you input the batch itself? So you would control the batch size I think? I'm also not familiar with...

I was playing around with a Word2Vec example to show off categorical inputs and it seems to me that maybe one reasonable solution is to do the subsetting within the...

The more I've thought about it, I think this is the way to go: 1. Each iteration generates and updates all gbm predictions 2. The data being fed in specifies...