adding support for LGBM_BoosterUpdateOneIterCustom #114
Conversation
src/wrapper.jl
Outdated
| end | ||
| numdata = LGBM_DatasetGetNumData(first(bst.datasets)) | ||
|
|
||
| if !(numdata == length(grads) == length(hessian)) |
There was a problem hiding this comment.
Should be equal to numdata * LGBM_BoosterNumModelPerIteration()
https://lightgbm.readthedocs.io/en/latest/C-API.html#c.LGBM_BoosterNumModelPerIteration
Please refer to microsoft/LightGBM#4815 (comment).
There was a problem hiding this comment.
@yaxxie I will wait for this before merging it
There was a problem hiding this comment.
Yep, I have some modifications to make WRT this.
|
|
||
| finished = LightGBM.LGBM_BoosterUpdateOneIterCustom(booster, randn(numdata), rand(numdata)) | ||
| pred1 = LightGBM.LGBM_BoosterGetPredict(booster, 0) | ||
| # check both types of float work |
There was a problem hiding this comment.
is this randn(numdata) type of Vector{<:AbstractFloat}?
There was a problem hiding this comment.
I think the test would fail with a MethodError if it was not
There was a problem hiding this comment.
it is typeof Vector{Float64} which is the same as the following test Float32.(randn(numdata))
There was a problem hiding this comment.
Well it's different -- the following test is Float32
The API takes only Float32 data, so if the user passes Float64, it needs to be converted. I test each way independently just to verify both entry points work as expected.
|
@FatemehTahavori I'm going to merge this now |
| """ | ||
| LGBM_BoosterUpdateOneIterCustom | ||
| Pass grads and 2nd derivatives corresponding to some custom loss function | ||
| grads and 2nd derivatives must be same cardinality as training data |
There was a problem hiding this comment.
as training data * number of trees per iteration
|
|
||
| if !((numdata*nummodels) == length(grads) == length(hessian)) | ||
| throw(DimensionMismatch( | ||
| "Gradients sizes ($(length(grads)), $(length(hessian))) don't match training data size ($numdata) * ($nummodels)" |
There was a problem hiding this comment.
number of trees per iteration * ($nummodels)
There was a problem hiding this comment.
Hmm? Not sure what you're expecting here, number of trees per iteration is the number of models, so the message would be like "doesnt match training data (22) * (3)" for example
|
|
||
| finished = LightGBM.LGBM_BoosterUpdateOneIterCustom(booster, randn(numdata*num_class), rand(numdata*num_class)) | ||
| pred1 = LightGBM.LGBM_BoosterGetPredict(booster, 0) | ||
| # check both types of float work |
As title