Rong Ou
7579905e18
Retry switching to per-thread default stream ( #9416 )
2023-07-26 07:09:12 +08:00
Jiaming Yuan
3a9996173e
Revert "Switch to per-thread default stream ( #9396 )" ( #9413 )
...
This reverts commit f7f673b00c
2023-07-24 12:03:28 -07:00
Jiaming Yuan
a196443a07
Implement sketching with Hessian on GPU. ( #9399 )
...
- Prepare for implementing approx on GPU.
- Unify the code path between weighted and uniform sketching on DMatrix.
2023-07-24 15:43:03 +08:00
Rong Ou
f7f673b00c
Switch to per-thread default stream ( #9396 )
2023-07-20 08:21:00 +08:00
Jiaming Yuan
d0916849a6
Remove unused weight from buffer for cat features. ( #9341 )
2023-07-04 01:07:09 +08:00
Jiaming Yuan
17fd3f55e9
Optimize adapter element counting on GPU. ( #9209 )
...
- Implement a simple `IterSpan` for passing iterators with size.
- Use shared memory for column size counts.
- Use one thread for each sample in row count to reduce atomic operations.
2023-05-30 23:28:43 +08:00
Jiaming Yuan
4d665b3fb0
Restore clang tidy test. ( #8861 )
2023-03-03 13:47:04 -08:00
Jiaming Yuan
16bca5d4a1
Support CPU input for device QuantileDMatrix. ( #8136 )
...
- Copy `GHistIndexMatrix` to `Ellpack` when needed.
2022-08-11 21:21:26 +08:00
Jiaming Yuan
8f8bd8147a
Fix LTR with weighted Quantile DMatrix. ( #7975 )
...
* Fix LTR with weighted Quantile DMatrix.
* Better tests.
2022-06-09 01:33:41 +08:00
Jiaming Yuan
ac9bfaa4f2
Handle missing values in dataframe with category dtype. ( #7331 )
...
* Replace -1 in pandas initializer.
* Unify `IsValid` functor.
* Mimic pandas data handling in cuDF glue code.
* Check invalid categories.
* Fix DDM sketching.
2021-10-28 03:33:54 +08:00
Jiaming Yuan
1c8fdf2218
Remove use of device_idx in dh::LaunchN. ( #7063 )
...
It's an unused parameter, removing it can make the CI log more readable.
2021-06-29 11:37:26 +08:00
Jiaming Yuan
7bcc8b3e5c
Use batched copy if. ( #6826 )
2021-04-06 10:34:04 +08:00
Jiaming Yuan
2241563f23
Handle duplicated values in sketching. ( #6178 )
...
* Accumulate weights in duplicated values.
* Fix device id in iterative dmatrix.
2020-10-10 19:32:44 +08:00
Jiaming Yuan
f0c63902ff
Use default allocator in sketching. ( #6182 )
2020-09-30 14:55:59 +08:00
Jiaming Yuan
e319b63f9e
Merge extract cuts into QuantileContainer. ( #6125 )
...
* Use pruning for initial summary construction.
2020-09-18 16:36:39 +08:00
Jiaming Yuan
ee70a2380b
Unify CPU hist sketching ( #5880 )
2020-08-12 01:33:06 +08:00
Jiaming Yuan
e471056ec4
Fix sketch size calculation. ( #5898 )
2020-07-17 08:33:16 +08:00
Jiaming Yuan
dd445af56e
Cleanup on device sketch. ( #5874 )
...
* Remove old functions.
* Merge weighted and un-weighted into a common interface.
2020-07-14 10:15:54 +08:00
Jiaming Yuan
048d969be4
Implement GK sketching on GPU. ( #5846 )
...
* Implement GK sketching on GPU.
* Strong tests on quantile building.
* Handle sparse dataset by binary searching the column index.
* Hypothesis test on dask.
2020-07-07 12:16:21 +08:00
Jiaming Yuan
1a0801238e
Implement iterative DMatrix. ( #5837 )
2020-07-03 11:44:52 +08:00
Jiaming Yuan
3028fa6b42
Implement weighted sketching for adapter. ( #5760 )
...
* Bounded memory tests.
* Fixed memory estimation.
2020-06-12 06:20:39 +08:00
Jiaming Yuan
e533908922
Expose device sketching in header. ( #5747 )
2020-06-02 13:02:53 +08:00
