[breaking] Use integer atomic for GPU histogram. (#7180)
On GPU we use rouding factor to truncate the gradient for deterministic results. This PR changes the gradient representation to fixed point number with exponent aligned with rounding factor.
[breaking] Drop non-deterministic histogram.
Use fixed point for shared memory.
This PR is to improve the performance of GPU Hist.
Co-authored-by: Andy Adinets <aadinets@nvidia.com>
This commit is contained in:
co-authored by
Andy Adinets
parent
e7d7ab6bc3
commit
7a1d67f9cb
@@ -1,8 +1,13 @@
|
||||
/*!
|
||||
* Copyright 2019-2021 by XGBoost Contributors
|
||||
*/
|
||||
#include <gtest/gtest.h>
|
||||
#include <vector>
|
||||
|
||||
#include <thrust/device_vector.h>
|
||||
#include <thrust/host_vector.h>
|
||||
#include <thrust/sequence.h>
|
||||
|
||||
#include "../../../../src/tree/gpu_hist/row_partitioner.cuh"
|
||||
#include "../../helpers.h"
|
||||
|
||||
|
||||
Reference in New Issue
Block a user