happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

convert : yield Mamba2Model/GraniteMoeModel modify_tensors (#19157) 

* convert : yield Mamba2Model/GraniteMoeModel modify_tensors

This commit updates the `GraniteHybridModel` class' modify_tensors
function to properly delegate to `Mamba2Model.modify_tensors` and
`GraniteMoeModel.modify_tensors` using 'yield from' instead of 'return'.

The motivation for this is that modify_tensors is a generator function
(it uses 'yield from'), but the two calls above use return statements
but don't yield anything which means that the the caller of this
function will not receive any yielded values from it. And this causes
layer tensors to be silently dropped during conversion.
This commit is contained in:
Daniel Bevenius 2026-01-28 16:49:36 +01:00 committed by GitHub
parent 0cd7032ca4
commit ebf5725870
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 6 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
convert_hf_to_gguf.py
View File

@ -8912,13 +8912,16 @@ class GraniteHybridModel(Mamba2Model, GraniteMoeModel):
name.endswith("block_sparse_moe.input_linear.weight")
or "shared_mlp" in name
):
return GraniteMoeModel.modify_tensors(self, data_torch, name, bid)
yield from GraniteMoeModel.modify_tensors(self, data_torch, name, bid)
return
# Determine whether this is a mamba layer or an attention layer
if bid in self._ssm_layers:
return Mamba2Model.modify_tensors(self, data_torch, name, bid)
yield from Mamba2Model.modify_tensors(self, data_torch, name, bid)
return
elif bid in self._attn_layers:
return GraniteMoeModel.modify_tensors(self, data_torch, name, bid)
yield from GraniteMoeModel.modify_tensors(self, data_torch, name, bid)
return
yield from ModelBase.modify_tensors(self, data_torch, name, bid)
def set_gguf_parameters(self):