Upload folder using huggingface_hub
Browse files
README.md
CHANGED
|
@@ -1,3 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# Capri
|
| 2 |
|
| 3 |
Capri is a compact image captioning model designed for high-throughput, plain-language descriptions.
|
|
@@ -82,3 +97,16 @@ Reasonable defaults:
|
|
| 82 |
- `decode_batch_size=1024`
|
| 83 |
|
| 84 |
On larger GPUs, decode often scales to `2048+`.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
tags:
|
| 6 |
+
- image-captioning
|
| 7 |
+
- multimodal
|
| 8 |
+
- vision-language
|
| 9 |
+
- qwen2
|
| 10 |
+
- siglip
|
| 11 |
+
datasets:
|
| 12 |
+
- merve/coco
|
| 13 |
+
pipeline_tag: image-to-text
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
# Capri
|
| 17 |
|
| 18 |
Capri is a compact image captioning model designed for high-throughput, plain-language descriptions.
|
|
|
|
| 97 |
- `decode_batch_size=1024`
|
| 98 |
|
| 99 |
On larger GPUs, decode often scales to `2048+`.
|
| 100 |
+
|
| 101 |
+
## Attribution
|
| 102 |
+
|
| 103 |
+
Trained on captions from the [COCO 2017](https://cocodataset.org/) dataset.
|
| 104 |
+
|
| 105 |
+
- Annotations © COCO Consortium, licensed under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)
|
| 106 |
+
- Images sourced from Flickr under their respective licenses; the dataset as a whole is not cleared for unrestricted commercial use
|
| 107 |
+
|
| 108 |
+
> Lin, T.-Y., et al. "Microsoft COCO: Common Objects in Context." ECCV 2014. [arXiv:1405.0312](https://arxiv.org/abs/1405.0312)
|
| 109 |
+
|
| 110 |
+
Built on top of:
|
| 111 |
+
- [Qwen/Qwen2.5-0.5B](https://huggingface.co/Qwen/Qwen2.5-0.5B) — Apache 2.0
|
| 112 |
+
- [google/siglip2-base-patch16-224](https://huggingface.co/google/siglip2-base-patch16-224) — Apache 2.0
|