Oguzhan Gencoglu
Oguzhan Gencoglu
I can do that quickly in code if I don't feel lazy but probably others would be interested.
+1 for a pypi package that anyone can install with `pip install`
@hahnec so what is the difference of quantization before conversion and quantizing the onnx model?
True. Good catch!
Any updates?
The probloem with one-hot encoding is Histogram Gradient boosting might learn weird interactions between each one-hot encoded feature during modeling. Therefore, it might not be the same as specifying that...
Same issue here. Any updates?
Is there any update to this?