(Noticed whilst reviewing #6695)

From the docs for numba.cuda.atomic.compare_and_swap:

class compare_and_swap(ary, old, val)

Conditionally assign val to the first element of an 1D array ary if the current value matches old.
Returns the current value as if it is loaded atomically.

It seem

Problem: the approximate method can still be slow for many trees
catboost version: master
Operating System: ubuntu 18.04
CPU: i9
GPU: RTX2080

Would be good to be able to specify how many trees to use for shapley. The model.predict and prediction_type versions allow this. lgbm/xgb allow this.

Currently, aggregation APIs (groupby, reductions, rolling, etc.) are scattered around in multiple files and there are inconsistencies between the directory structures in cpp/include/, cpp/src/, cpp/tests/, and cpp/benchmarks/. For example:

cpp/include/:

• include/cudf/aggregation.hpp
• include/cudf/groupby.hpp
• include/cudf/rolling.hpp
• ....

cpp/src/:

• src/aggregati

Current implementation of join can be improved by performing the operation in a single call to the backend kernel instead of multiple calls.

This is a fairly easy kernel and may be a good issue for someone getting to know CUDA/ArrayFire internals. Ping me if you want additional info.

PR NVIDIA/cub#218 fixes this CUB's radix sort. We should:

• Check whether Thrust's other backends handle this case correctly.
• Provide a guarantee of this in the stable_sort documentation.
• Add regression tests to enforce this on all backends.

confusion_matrix should automatically convert dtypes as appropriate in order to avoid failing, like other metric functions.

from sklearn.metrics import confusion_matrix
import numpy as np
import cuml
​
y = np.array([0.0, 1.0, 0.0])
y_pred = np.array([0.0, 1.0, 1.0])
print(confusion_matrix(y, y_pred))
cuml.metrics.confusion_matrix(y, y_pred)
[[1 1]
 [0 1]]
----------------

I often use -v just to see that something is going on, but a progress bar (enabled by default) would serve the same purpose and be more concise.

We can just factor out the code from futhark bench for this.

Thank you for this fantastic work!

Could it be possible the fit_transform() method returns the KL divergence of the run?

Thx!