christian.schwarz/DeepSeek-R1-Zero
1
0
Fork 0
mirror of https://huggingface.co/deepseek-ai/DeepSeek-R1-Zero synced 2026-03-11 05:26:30 +01:00
Code Issues Packages Projects Releases Wiki Activity

Remove unavailable import

Browse Source 
The Transformers codebase doesn't really support Torch 1.x anymore, so the compatibility checks have been removed in the most recent versions!
This commit is contained in:
Matthew Carrigan 2025-01-27 23:40:48 +00:00 committed by system
parent d9c339cd7d
commit 03f6b95966
No known key found for this signature in database
GPG Key ID: 6A528E38E0733467
1 changed files with 0 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
modeling_deepseek.py
View File

@ -43,7 +43,6 @@ from transformers.modeling_outputs import (
from transformers.modeling_utils import PreTrainedModel from transformers.modeling_utils import PreTrainedModel
from transformers.pytorch_utils import ( from transformers.pytorch_utils import (
ALL_LAYERNORM_LAYERS, ALL_LAYERNORM_LAYERS,
is_torch_greater_or_equal_than_1_13,
) )
from transformers.utils import ( from transformers.utils import (
add_start_docstrings, add_start_docstrings,
@ -66,9 +65,6 @@ if is_flash_attn_2_available():
# This makes `_prepare_4d_causal_attention_mask` a leaf function in the FX graph. # This makes `_prepare_4d_causal_attention_mask` a leaf function in the FX graph.
# It means that the function will not be traced through and simply appear as a node in the graph. # It means that the function will not be traced through and simply appear as a node in the graph.
if is_torch_fx_available(): if is_torch_fx_available():
if not is_torch_greater_or_equal_than_1_13:
import torch.fx
_prepare_4d_causal_attention_mask = torch.fx.wrap(_prepare_4d_causal_attention_mask) _prepare_4d_causal_attention_mask = torch.fx.wrap(_prepare_4d_causal_attention_mask)