minor
Browse files
README.md
CHANGED
|
@@ -29,4 +29,23 @@ https://github.com/baaivision/EVA/tree/master/EVA-CLIP
|
|
| 29 |
- To construct Merged-2B, we merged 1.6 billion samples from [LAION-2B](https://laion.ai/blog/laion-5b/) dataset with 0.4 billion samples from [COYO-700M](https://github.com/kakaobrain/coyo-dataset).
|
| 30 |
|
| 31 |
- To our knowledge, EVA-CLIP series are the most performant open-sourced CLIP models at all scales, evaluated via zero-shot classification performance, especially on mainstream classification benchmarks such as ImageNet along with its variants.
|
| 32 |
-
For more details about EVA-CLIP, please refer to our [paper (coming very soon)]().
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
- To construct Merged-2B, we merged 1.6 billion samples from [LAION-2B](https://laion.ai/blog/laion-5b/) dataset with 0.4 billion samples from [COYO-700M](https://github.com/kakaobrain/coyo-dataset).
|
| 30 |
|
| 31 |
- To our knowledge, EVA-CLIP series are the most performant open-sourced CLIP models at all scales, evaluated via zero-shot classification performance, especially on mainstream classification benchmarks such as ImageNet along with its variants.
|
| 32 |
+
For more details about EVA-CLIP, please refer to our [paper (coming very soon)]().
|
| 33 |
+
|
| 34 |
+
### pretrained
|
| 35 |
+
<div align="center">
|
| 36 |
+
|
| 37 |
+
| model name | total #params | training precision | download link |
|
| 38 |
+
|:-----------|:------:|:------:|:------:|
|
| 39 |
+
| `EVA01_g_psz14` | 1.0B | `fp16` | [🤗 HF link](https://huggingface.co/QuanSun/EVA-CLIP/blob/main/EVA01_g_psz14.pt) (`2.0GB`) |
|
| 40 |
+
| `EVA02_B_psz14to16` | 86M | `fp16` | [🤗 HF link](https://huggingface.co/QuanSun/EVA-CLIP/blob/main/EVA02_B_psz14to16.pt) (`176MB`) |
|
| 41 |
+
| `EVA02_L_psz14` | 304M | `fp16` | [🤗 HF link](https://huggingface.co/QuanSun/EVA-CLIP/blob/main/EVA02_L_psz14.pt) (`609MB`) |
|
| 42 |
+
| `EVA02_CLIP_L_psz14_224to336` | 428M | `fp16` | [🤗 HF link](https://huggingface.co/QuanSun/EVA-CLIP/blob/main/EVA02_CLIP_L_psz14_224to336.pt) (`857MB`) |
|
| 43 |
+
| `EVA02_E_psz14` | 4.4B | `fp16` | [🤗 HF link](https://huggingface.co/QuanSun/EVA-CLIP/blob/main/EVA02_E_psz14.pt) (`8.7GB`) |
|
| 44 |
+
| `openai/clip-vit-base-patch16`| 149M | `fp16` | [🤗 HF link](https://huggingface.co/openai/clip-vit-base-patch16/blob/main/pytorch_model.bin) (`599MB`) |
|
| 45 |
+
| `openai/clip-vit-large-patch14`| 428M | `fp16` | [🤗 HF link](https://huggingface.co/openai/clip-vit-large-patch14/blob/main/pytorch_model.bin) (`1.7GB`) |
|
| 46 |
+
| `laion/CLIP-ViT-H-14-laion2B-s32B-b79K`| 1.0B | `bf16` | [🤗 HF link](https://huggingface.co/laion/CLIP-ViT-H-14-laion2B-s32B-b79K/blob/main/pytorch_model.bin) (`3.9GB`) |
|
| 47 |
+
|
| 48 |
+
</div>
|
| 49 |
+
EVA02_B_psz14to16 interpolates the kernel size of patch_embed from 14x14 to 16x16, and interpolate the pos_embed from 16x16 to 14x14.
|
| 50 |
+
|
| 51 |
+
EVA02_CLIP_L_psz14_224to336 interpolates the pos_embed from 16x16 to 24x24 for training EVA02_CLIP_L_336_psz14_s6B.
|