ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Triton windows

Browse source
This commit is contained in:
Daniel Han 2024-12-24 00:17:33 -08:00
commit 77ca6e2dca
3 changed files with 9 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

7
pyproject.toml
View file

@ -32,6 +32,12 @@ include-package-data = false
exclude = ["images*"]
[project.optional-dependencies]
triton = [
"triton @ https://github.com/woct0rdho/triton-windows/releases/download/v3.1.0-windows.post5/triton-3.1.0-cp39-cp39-win_amd64.whl ; python_version=='3.9' and platform_system == 'Windows'",
"triton @ https://github.com/woct0rdho/triton-windows/releases/download/v3.1.0-windows.post5/triton-3.1.0-cp310-cp310-win_amd64.whl ; python_version=='3.10' and platform_system == 'Windows'",
"triton @ https://github.com/woct0rdho/triton-windows/releases/download/v3.1.0-windows.post5/triton-3.1.0-cp311-cp311-win_amd64.whl ; python_version=='3.11' and platform_system == 'Windows'",
"triton @ https://github.com/woct0rdho/triton-windows/releases/download/v3.1.0-windows.post5/triton-3.1.0-cp312-cp312-win_amd64.whl ; python_version=='3.12' and platform_system == 'Windows'",
]
huggingface = [
"unsloth_zoo>=2024.12.3",
"packaging",
@ -49,6 +55,7 @@ huggingface = [
"protobuf<4.0.0",
"huggingface_hub",
"hf_transfer",
"unsloth[triton]",
]
cu118only = [
"xformers @ https://download.pytorch.org/whl/cu118/xformers-0.0.22.post7%2Bcu118-cp39-cp39-manylinux2014_x86_64.whl ; python_version=='3.9' and platform_system == 'Linux'",

2
unsloth/models/gemma.py
View file

@ -77,7 +77,7 @@ pass
def GemmaDecoderLayer_fast_forward(
self,
hidden_states: torch.Tensor,
causal_mask: Optional[xformers.attn_bias.BlockDiagonalCausalMask] = None,
causal_mask: Optional[BlockDiagonalCausalMask] = None,
attention_mask: Optional[torch.Tensor] = None,
position_ids: Optional[torch.LongTensor] = None,
past_key_value: Optional[Tuple[torch.Tensor]] = None,

2
unsloth/models/gemma2.py
View file

@ -75,7 +75,7 @@ pass
def Gemma2Attention_fast_forward(
self,
hidden_states: torch.Tensor,
causal_mask: Optional[xformers.attn_bias.BlockDiagonalCausalMask] = None,
causal_mask: Optional[BlockDiagonalCausalMask] = None,
attention_mask: Optional[torch.Tensor] = None,
position_ids: Optional[torch.LongTensor] = None,
past_key_value: Optional[Tuple[torch.Tensor]] = None,