From: Thérence <redacted>
Date: Mon, 18 Mar 2024 08:17:00 +0000 (+0100)
Subject: convert : add support for CamembertModel architecture (#6119)
X-Git-Tag: upstream/0.0.4488~2035
X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=9b03719ad712e2dc36c5c0c20f352bf3e4bda332;p=pkg%2Fggml%2Fsources%2Fllama.cpp

convert : add support for CamembertModel architecture (#6119)

Adding support for CamembertModel architecture used by :
https://huggingface.co/dangvantuan/sentence-camembert-large
---

diff --git a/convert-hf-to-gguf.py b/convert-hf-to-gguf.py
index cf1f98d6..1e49d56c 100755
--- a/convert-hf-to-gguf.py
+++ b/convert-hf-to-gguf.py
@@ -1634,7 +1634,7 @@ in chat mode so that the conversation can end normally.")
                 self.post_write_tensors(tensor_map, name, data_torch)
 
 
-@Model.register("BertModel")
+@Model.register("BertModel", "CamembertModel")
 class BertModel(Model):
     model_arch = gguf.MODEL_ARCH.BERT