warpgbm 0.1.24__tar.gz → 0.1.25__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {warpgbm-0.1.24/warpgbm.egg-info → warpgbm-0.1.25}/PKG-INFO +5 -1
- {warpgbm-0.1.24 → warpgbm-0.1.25}/README.md +4 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/pyproject.toml +1 -1
- warpgbm-0.1.25/version.txt +1 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25/warpgbm.egg-info}/PKG-INFO +5 -1
- warpgbm-0.1.24/version.txt +0 -1
- {warpgbm-0.1.24 → warpgbm-0.1.25}/LICENSE +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/MANIFEST.in +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/setup.cfg +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/setup.py +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/tests/__init__.py +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/tests/numerai_test.py +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/tests/test_fit_predict_corr.py +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/warpgbm/__init__.py +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/warpgbm/core.py +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/warpgbm/cuda/__init__.py +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/warpgbm/cuda/best_split_kernel.cu +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/warpgbm/cuda/binner.cu +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/warpgbm/cuda/histogram_kernel.cu +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/warpgbm/cuda/node_kernel.cpp +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/warpgbm/cuda/predict.cu +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/warpgbm.egg-info/SOURCES.txt +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/warpgbm.egg-info/dependency_links.txt +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/warpgbm.egg-info/requires.txt +0 -0
- {warpgbm-0.1.24 → warpgbm-0.1.25}/warpgbm.egg-info/top_level.txt +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.4
|
|
2
2
|
Name: warpgbm
|
|
3
|
-
Version: 0.1.
|
|
3
|
+
Version: 0.1.25
|
|
4
4
|
Summary: A fast GPU-accelerated Gradient Boosted Decision Tree library with PyTorch + CUDA
|
|
5
5
|
License: GNU GENERAL PUBLIC LICENSE
|
|
6
6
|
Version 3, 29 June 2007
|
|
@@ -877,6 +877,7 @@ No installation required — just press **"Open in Playground"**, then **Run All
|
|
|
877
877
|
- `threads_per_block`: CUDA threads per block (default: 32)
|
|
878
878
|
- `rows_per_thread`: Number of training rows processed per thread (default: 4)
|
|
879
879
|
- `L2_reg`: L2 regularizer (default: 1e-6)
|
|
880
|
+
- `colsample_bytree`: Proportion of features to subsample to grow each tree (default: 1)
|
|
880
881
|
|
|
881
882
|
### Methods:
|
|
882
883
|
```
|
|
@@ -918,3 +919,6 @@ WarpGBM builds on the shoulders of PyTorch, scikit-learn, LightGBM, and the CUDA
|
|
|
918
919
|
|
|
919
920
|
- Adjust gain in split kernel and added support for an eval set with early stopping based on MSE.
|
|
920
921
|
|
|
922
|
+
### v0.1.25
|
|
923
|
+
|
|
924
|
+
- Added `colsample_bytree` parameter and new test using Numerai data.
|
|
@@ -189,6 +189,7 @@ No installation required — just press **"Open in Playground"**, then **Run All
|
|
|
189
189
|
- `threads_per_block`: CUDA threads per block (default: 32)
|
|
190
190
|
- `rows_per_thread`: Number of training rows processed per thread (default: 4)
|
|
191
191
|
- `L2_reg`: L2 regularizer (default: 1e-6)
|
|
192
|
+
- `colsample_bytree`: Proportion of features to subsample to grow each tree (default: 1)
|
|
192
193
|
|
|
193
194
|
### Methods:
|
|
194
195
|
```
|
|
@@ -230,3 +231,6 @@ WarpGBM builds on the shoulders of PyTorch, scikit-learn, LightGBM, and the CUDA
|
|
|
230
231
|
|
|
231
232
|
- Adjust gain in split kernel and added support for an eval set with early stopping based on MSE.
|
|
232
233
|
|
|
234
|
+
### v0.1.25
|
|
235
|
+
|
|
236
|
+
- Added `colsample_bytree` parameter and new test using Numerai data.
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
0.1.25
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.4
|
|
2
2
|
Name: warpgbm
|
|
3
|
-
Version: 0.1.
|
|
3
|
+
Version: 0.1.25
|
|
4
4
|
Summary: A fast GPU-accelerated Gradient Boosted Decision Tree library with PyTorch + CUDA
|
|
5
5
|
License: GNU GENERAL PUBLIC LICENSE
|
|
6
6
|
Version 3, 29 June 2007
|
|
@@ -877,6 +877,7 @@ No installation required — just press **"Open in Playground"**, then **Run All
|
|
|
877
877
|
- `threads_per_block`: CUDA threads per block (default: 32)
|
|
878
878
|
- `rows_per_thread`: Number of training rows processed per thread (default: 4)
|
|
879
879
|
- `L2_reg`: L2 regularizer (default: 1e-6)
|
|
880
|
+
- `colsample_bytree`: Proportion of features to subsample to grow each tree (default: 1)
|
|
880
881
|
|
|
881
882
|
### Methods:
|
|
882
883
|
```
|
|
@@ -918,3 +919,6 @@ WarpGBM builds on the shoulders of PyTorch, scikit-learn, LightGBM, and the CUDA
|
|
|
918
919
|
|
|
919
920
|
- Adjust gain in split kernel and added support for an eval set with early stopping based on MSE.
|
|
920
921
|
|
|
922
|
+
### v0.1.25
|
|
923
|
+
|
|
924
|
+
- Added `colsample_bytree` parameter and new test using Numerai data.
|
warpgbm-0.1.24/version.txt
DELETED
|
@@ -1 +0,0 @@
|
|
|
1
|
-
0.1.24
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|