CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-11 10:36:54 +00:00
Code Issues Packages Projects Releases Wiki Activity

llama : add support for BertForSequenceClassification reranker (#13858)

Browse Source 
* convert: add support for BertForSequenceClassification

* add support for reranking using BertForSequenceClassification

* merge checks of eos and sep

* fix lint

---------

Co-authored-by: dinhhuy <huy.dinh@brains-tech.co.jp>
This commit is contained in:
Đinh Trọng Huy
2025-05-29 02:01:58 +09:00
committed by GitHub
parent aa6dff05be
commit e0e3aa231d
4 changed files with 42 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
convert_hf_to_gguf.py
View File

@@ -3682,7 +3682,7 @@ class InternLM3Model(TextModel):
return [(self.map_tensor_name(name), data_torch)]
@ModelBase.register("BertModel", "BertForMaskedLM", "CamembertModel")
@ModelBase.register("BertModel", "BertForMaskedLM", "CamembertModel", "BertForSequenceClassification")
class BertModel(TextModel):
model_arch = gguf.MODEL_ARCH.BERT
@@ -3745,6 +3745,13 @@ class BertModel(TextModel):
if name.startswith("cls.seq_relationship"):
return []
# For BertForSequenceClassification (direct projection layer)
if name == "classifier.weight":
name = "classifier.out_proj.weight"
if name == "classifier.bias":
name = "classifier.out_proj.bias"
return [(self.map_tensor_name(name), data_torch)]
def _xlmroberta_tokenizer_init(self) -> None: