Skip to content

Commit 7740c6e

Browse files
committed
Fixed url in setup.py (#38), updated changelog.
1 parent 439f2b0 commit 7740c6e

File tree

2 files changed

+13
-2
lines changed

2 files changed

+13
-2
lines changed

CHANGELOG.md

Lines changed: 11 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -106,3 +106,14 @@ Bug fixes:
106106
- fixed an import of bnb.utils 2e630b55f51d454f3bd723dffda68a07ef93190c
107107

108108
We thank @mryab, @mbrukman, @chessgecko, @dbaranchuk for pull request with bug fixes and new features.
109+
110+
111+
### 0.34.0
112+
113+
#### Bug fixes and memory efficient backprop
114+
115+
Features:
116+
- Linear8bitLt layer now supports `memory_efficient_backward=True` which enables backprop of gradients through frozen weights.
117+
118+
Bug fixes:
119+
- fixed an issue where too many threads were created in blockwise quantization on the CPU for large tensors

setup.py

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -18,13 +18,13 @@ def read(fname):
1818

1919
setup(
2020
name=f"bitsandbytes",
21-
version=f"0.33.1",
21+
version=f"0.34.0",
2222
author="Tim Dettmers",
2323
author_email="[email protected]",
2424
description="8-bit optimizers and matrix multiplication routines.",
2525
license="MIT",
2626
keywords="gpu optimizers optimization 8-bit quantization compression",
27-
url="http://packages.python.org/bitsandbytes",
27+
url="https://github.com/TimDettmers/bitsandbytes",
2828
packages=find_packages(),
2929
entry_points={
3030
"console_scripts": ["debug_cuda = bitsandbytes.debug_cli:cli"],

0 commit comments

Comments
 (0)