vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

llava : update readme

Browse source
This commit is contained in:
Georgi Gerganov 2024-02-13 19:59:00 +02:00
parent 6b8d69b451
commit a2848854a4
No known key found for this signature in database
GPG key ID: 449E073F9DC10735
2 changed files with 9 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

6
examples/llava/README.md
View file

@ -49,8 +49,12 @@ python ./convert.py ../llava-v1.5-7b
Now both the LLaMA part and the image encoder is in the `llava-v1.5-7b` directory.
## LLaVA 1.6
- Use `llava-surgery-v2.py`
## TODO
- [ ] Support non-CPU backend for the image encoding part.
- [x] Support non-CPU backend for the image encoding part.
- [ ] Support different sampling methods.
- [ ] Support more model variants.

8
examples/llava/convert-image-encoder-to-gguf.py
View file

@ -78,9 +78,9 @@ ap.add_argument("--text-only", action="store_true", required=False,
help="Save a text-only model. It can't be used to encode images")
ap.add_argument("--vision-only", action="store_true", required=False,
help="Save a vision-only model. It can't be used to encode texts")
ap.add_argument("--clip_model_is_vision", action="store_true", required=False,
ap.add_argument("--clip-model-is-vision", action="store_true", required=False,
help="The clip model is a pure vision model (ShareGPT4V vision extract for example)")
ap.add_argument("--clip_model_is_openclip", action="store_true", required=False,
ap.add_argument("--clip-model-is-openclip", action="store_true", required=False,
help="The clip model is from openclip (for ViT-SO400M type))")
ap.add_argument("--llava-projector", help="Path to llava.projector file. If specified, save an image encoder for LLaVA models.")
ap.add_argument("--projector-type", help="Type of projector. Possible values: mlp, ldp", choices=["mlp", "ldp"], default="mlp")
@ -89,8 +89,8 @@ ap.add_argument("-o", "--output-dir", help="Directory to save GGUF files. Defaul
# Example --image_mean 0.5 0.5 0.5 --image_std 0.5 0.5 0.5
default_image_mean = [0.48145466, 0.4578275, 0.40821073]
default_image_std = [0.26862954, 0.26130258, 0.27577711]
ap.add_argument('--image_mean', type=float, nargs='+', help='Mean of the images for normalization (overrides processor) ', default=None)
ap.add_argument('--image_std', type=float, nargs='+', help='Standard deviation of the images for normalization (overrides processor)', default=None)
ap.add_argument('--image-mean', type=float, nargs='+', help='Mean of the images for normalization (overrides processor) ', default=None)
ap.add_argument('--image-std', type=float, nargs='+', help='Standard deviation of the images for normalization (overrides processor)', default=None)
# with proper
args = ap.parse_args()