Add library name and paper link to model card
Browse filesThis PR adds the `library_name` to the metadata section of the model card, clarifying that this model works with the Hugging Face `transformers` library. It also adds a link to the paper on Hugging Face.
README.md
CHANGED
|
@@ -1,11 +1,13 @@
|
|
| 1 |
---
|
| 2 |
license: llama2
|
| 3 |
pipeline_tag: video-text-to-text
|
|
|
|
| 4 |
---
|
|
|
|
| 5 |
# Chat-UniVi: Unified Visual Representation Empowers Large Language Models with Image and Video Understanding
|
| 6 |
|
| 7 |
**Paper or resources for more information:**
|
| 8 |
-
[[Paper](https://huggingface.co/papers/
|
| 9 |
|
| 10 |
## License
|
| 11 |
Llama 2 is licensed under the LLAMA 2 Community License,
|
|
@@ -133,9 +135,11 @@ if __name__ == '__main__':
|
|
| 133 |
|
| 134 |
cur_prompt = qs
|
| 135 |
if model.config.mm_use_im_start_end:
|
| 136 |
-
qs = DEFAULT_IM_START_TOKEN + DEFAULT_IMAGE_TOKEN * slice_len + DEFAULT_IM_END_TOKEN + '
|
|
|
|
| 137 |
else:
|
| 138 |
-
qs = DEFAULT_IMAGE_TOKEN * slice_len + '
|
|
|
|
| 139 |
|
| 140 |
conv = conv_templates[conv_mode].copy()
|
| 141 |
conv.append_message(conv.roles[0], qs)
|
|
@@ -224,9 +228,11 @@ if __name__ == '__main__':
|
|
| 224 |
if image_path is not None:
|
| 225 |
cur_prompt = qs
|
| 226 |
if model.config.mm_use_im_start_end:
|
| 227 |
-
qs = DEFAULT_IM_START_TOKEN + DEFAULT_IMAGE_TOKEN + DEFAULT_IM_END_TOKEN + '
|
|
|
|
| 228 |
else:
|
| 229 |
-
qs = DEFAULT_IMAGE_TOKEN + '
|
|
|
|
| 230 |
|
| 231 |
conv = conv_templates[conv_mode].copy()
|
| 232 |
conv.append_message(conv.roles[0], qs)
|
|
|
|
| 1 |
---
|
| 2 |
license: llama2
|
| 3 |
pipeline_tag: video-text-to-text
|
| 4 |
+
library_name: transformers
|
| 5 |
---
|
| 6 |
+
|
| 7 |
# Chat-UniVi: Unified Visual Representation Empowers Large Language Models with Image and Video Understanding
|
| 8 |
|
| 9 |
**Paper or resources for more information:**
|
| 10 |
+
[[Paper](https://huggingface.co/papers/2503.04504)] [[Code](https://github.com/PKU-YuanGroup/Chat-UniVi)]
|
| 11 |
|
| 12 |
## License
|
| 13 |
Llama 2 is licensed under the LLAMA 2 Community License,
|
|
|
|
| 135 |
|
| 136 |
cur_prompt = qs
|
| 137 |
if model.config.mm_use_im_start_end:
|
| 138 |
+
qs = DEFAULT_IM_START_TOKEN + DEFAULT_IMAGE_TOKEN * slice_len + DEFAULT_IM_END_TOKEN + '
|
| 139 |
+
' + qs
|
| 140 |
else:
|
| 141 |
+
qs = DEFAULT_IMAGE_TOKEN * slice_len + '
|
| 142 |
+
' + qs
|
| 143 |
|
| 144 |
conv = conv_templates[conv_mode].copy()
|
| 145 |
conv.append_message(conv.roles[0], qs)
|
|
|
|
| 228 |
if image_path is not None:
|
| 229 |
cur_prompt = qs
|
| 230 |
if model.config.mm_use_im_start_end:
|
| 231 |
+
qs = DEFAULT_IM_START_TOKEN + DEFAULT_IMAGE_TOKEN + DEFAULT_IM_END_TOKEN + '
|
| 232 |
+
' + qs
|
| 233 |
else:
|
| 234 |
+
qs = DEFAULT_IMAGE_TOKEN + '
|
| 235 |
+
' + qs
|
| 236 |
|
| 237 |
conv = conv_templates[conv_mode].copy()
|
| 238 |
conv.append_message(conv.roles[0], qs)
|