vrr/kvcache-ai-ktransformers
2
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2025-09-09 13:55:27 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

support KExpertsMarlin backend

Browse source
This commit is contained in:
Azure 2025-02-07 05:57:40 +00:00
parent 0262f954c7
commit c4d9bc6670
5 changed files with 214 additions and 46 deletions
Download patch file Download diff file Expand all files Collapse all files

4
ktransformers/operators/linear.py
View file

@ -138,10 +138,10 @@ class KLinearTorch(KLinearBase):
if w is None: w = self.load_weight(device=device)
if isinstance(w, nn.Parameter):
self.w = w.to(dtype=self.dtype).view(self.out_features, self.in_features).T
self.w = w.to(dtype=self.dtype).T
self.has_bias = False
elif isinstance(w, tuple):
self.w = w[0].to(dtype=self.dtype).view(self.out_features, self.in_features).T
self.w = w[0].to(dtype=self.dtype).T
self.bias = w[1].to(dtype=self.dtype)
self.has_bias = True
else: