Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
library_name: transformers
|
| 5 |
+
---
|
| 6 |
+
This is the LLaVa model in Huggingface format which was derived from [actuaryzhang/product-taxonomy-llava-v1.6-13b](https://huggingface.co/actuaryzhang/product-taxonomy-llava-v1.6-13b)
|
| 7 |
+
|
| 8 |
+
A modified version of the https://github.com/huggingface/transformers/blob/0f67ba1d741d65b07d549daf4ee157609ce4f9c1/src/transformers/models/llava/convert_llava_weights_to_hf.py#L1 for the OSS to Huggingface format was used to create this model.
|
| 9 |
+
|
| 10 |
+
Note that though this model is marked as v1.6 (LLaVa-NeXT), it doesn't completely inherit the improvements on of LLaVa-NeXT. In particular, it does not leverage LLaVa-NeXT's improvements on image processing, and the `image_newline` tensor is missing
|