root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-24 02:14:35 +00:00
Code Issues Actions 11 Packages Projects Releases Wiki Activity

llava : avoid changing the original BakLLaVA model (#5577)

Browse Source 
This is a follup of Commit fc0c8d286a
("llava : update surgery script to not remove tensors") but this time
the change is to the BakLLaVA specific part of the surgery script.

I've been able to test this using SkunkworksAI/BakLLaVA-1 and it works
as expected using the instructions in README.md.

Signed-off-by: Daniel Bevenius <daniel.bevenius@gmail.com>
This commit is contained in:
Daniel Bevenius 2024-02-19 09:31:59 +01:00 committed by GitHub
parent 4480542b22
commit 7084755396
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 0 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
examples/llava/llava-surgery.py
View File

@ -25,9 +25,6 @@ if len(clip_tensors) > 0:
clip = {name.replace("vision_tower.vision_tower.", ""): checkpoint[name].float() for name in clip_tensors}
torch.save(clip, f"{args.model}/llava.clip")
# remove these tensors
for name in clip_tensors:
del checkpoint[name]
# added tokens should be removed to be able to convert Mistral models
if os.path.exists(f"{args.model}/added_tokens.json"):
@ -35,7 +32,6 @@ if len(clip_tensors) > 0:
f.write("{}\n")
torch.save(checkpoint, path)
print("Done!")
print(f"Now you can convert {args.model} to a regular LLaMA GGUF file.")