Upload 11 files
Browse files- README.md +46 -0
- added_tokens.json +4 -0
- config.json +28 -0
- merges.txt +0 -0
- metaclip_h14_fullcc2.5b.bin +3 -0
- model.safetensors +3 -0
- preprocessor_config.json +28 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +6 -0
- tokenizer_config.json +33 -0
- vocab.json +0 -0
README.md
ADDED
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cc-by-nc-4.0
|
3 |
+
tags:
|
4 |
+
- vision
|
5 |
+
- metaclip
|
6 |
+
widget:
|
7 |
+
- src: https://huggingface.co/datasets/mishig/sample_images/resolve/main/cat-dog-music.png
|
8 |
+
candidate_labels: playing music, playing sports
|
9 |
+
example_title: Cat & Dog
|
10 |
+
---
|
11 |
+
|
12 |
+
# MetaCLIP model, huge-sized version, patch resolution 14
|
13 |
+
|
14 |
+
MetaCLIP model applied to 2.5 billion data points of CommonCrawl (CC). It was introduced in the paper [Demystifying CLIP Data](https://arxiv.org/abs/2309.16671) by Xu et al. and first released in [this repository](https://github.com/facebookresearch/MetaCLIP).
|
15 |
+
|
16 |
+
Disclaimer: The team releasing MetaCLIP did not write a model card for this model so this model card has been written by the Hugging Face team.
|
17 |
+
|
18 |
+
## Model description
|
19 |
+
|
20 |
+
The [Demystifying CLIP Data](https://arxiv.org/abs/2309.16671) paper aims to reveal CLIP’s method around training data curation. OpenAI never open-sourced code regarding their data preparation pipeline.
|
21 |
+
|
22 |
+
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/clip_overview.jpg"
|
23 |
+
alt="drawing" width="600"/>
|
24 |
+
|
25 |
+
<small> CLIP high-level overview. Taken from the <a href="https://arxiv.org/abs/2103.00020">CLIP paper</a>. </small>
|
26 |
+
|
27 |
+
## Intended uses & limitations
|
28 |
+
|
29 |
+
You can use the raw model for linking images with text in a shared embedding space. This enables things like zero-shot image classification, text-based image retrieval, image-based text retrieval, etc.
|
30 |
+
|
31 |
+
### How to use
|
32 |
+
|
33 |
+
We refer to the [docs](https://huggingface.co/docs/transformers/main/en/model_doc/clip#usage). Just replace the names of the models on the hub.
|
34 |
+
|
35 |
+
### BibTeX entry and citation info
|
36 |
+
|
37 |
+
```bibtex
|
38 |
+
@misc{xu2023demystifying,
|
39 |
+
title={Demystifying CLIP Data},
|
40 |
+
author={Hu Xu and Saining Xie and Xiaoqing Ellen Tan and Po-Yao Huang and Russell Howes and Vasu Sharma and Shang-Wen Li and Gargi Ghosh and Luke Zettlemoyer and Christoph Feichtenhofer},
|
41 |
+
year={2023},
|
42 |
+
eprint={2309.16671},
|
43 |
+
archivePrefix={arXiv},
|
44 |
+
primaryClass={cs.CV}
|
45 |
+
}
|
46 |
+
```
|
added_tokens.json
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<|endoftext|>": 49407,
|
3 |
+
"<|startoftext|>": 49406
|
4 |
+
}
|
config.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"CLIPModel"
|
4 |
+
],
|
5 |
+
"initializer_factor": 1.0,
|
6 |
+
"logit_scale_init_value": 2.6592,
|
7 |
+
"model_type": "clip",
|
8 |
+
"projection_dim": 1024,
|
9 |
+
"text_config": {
|
10 |
+
"hidden_size": 1024,
|
11 |
+
"intermediate_size": 4096,
|
12 |
+
"model_type": "clip_text_model",
|
13 |
+
"num_attention_heads": 16,
|
14 |
+
"num_hidden_layers": 24,
|
15 |
+
"projection_dim": 1024
|
16 |
+
},
|
17 |
+
"torch_dtype": "float32",
|
18 |
+
"transformers_version": "4.34.0",
|
19 |
+
"vision_config": {
|
20 |
+
"hidden_size": 1280,
|
21 |
+
"intermediate_size": 5120,
|
22 |
+
"model_type": "clip_vision_model",
|
23 |
+
"num_attention_heads": 16,
|
24 |
+
"num_hidden_layers": 32,
|
25 |
+
"patch_size": 14,
|
26 |
+
"projection_dim": 1024
|
27 |
+
}
|
28 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
metaclip_h14_fullcc2.5b.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b790ce09d92080471aaa8718293126169f9d07b8147b9677bb2f14ddf8d9fff4
|
3 |
+
size 3944704310
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6173e1a29c7b449ce5ebe788936625c909b4d9059eb89bc4ec80bbdc00bc6571
|
3 |
+
size 3944549372
|
preprocessor_config.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"crop_size": {
|
3 |
+
"height": 224,
|
4 |
+
"width": 224
|
5 |
+
},
|
6 |
+
"do_center_crop": true,
|
7 |
+
"do_convert_rgb": true,
|
8 |
+
"do_normalize": true,
|
9 |
+
"do_rescale": true,
|
10 |
+
"do_resize": true,
|
11 |
+
"image_mean": [
|
12 |
+
0.48145466,
|
13 |
+
0.4578275,
|
14 |
+
0.40821073
|
15 |
+
],
|
16 |
+
"image_processor_type": "CLIPImageProcessor",
|
17 |
+
"image_std": [
|
18 |
+
0.26862954,
|
19 |
+
0.26130258,
|
20 |
+
0.27577711
|
21 |
+
],
|
22 |
+
"processor_class": "CLIPProcessor",
|
23 |
+
"resample": 3,
|
24 |
+
"rescale_factor": 0.00392156862745098,
|
25 |
+
"size": {
|
26 |
+
"shortest_edge": 224
|
27 |
+
}
|
28 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1930613c95917df0af9d07d368a20c68a181bc9a8821ac145648f0f7fcc095e5
|
3 |
+
size 3944739266
|
special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<|startoftext|>",
|
3 |
+
"eos_token": "<|endoftext|>",
|
4 |
+
"pad_token": "<|endoftext|>",
|
5 |
+
"unk_token": "<|endoftext|>"
|
6 |
+
}
|
tokenizer_config.json
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"added_tokens_decoder": {
|
4 |
+
"49406": {
|
5 |
+
"content": "<|startoftext|>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": true,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false,
|
10 |
+
"special": false
|
11 |
+
},
|
12 |
+
"49407": {
|
13 |
+
"content": "<|endoftext|>",
|
14 |
+
"lstrip": false,
|
15 |
+
"normalized": false,
|
16 |
+
"rstrip": false,
|
17 |
+
"single_word": false,
|
18 |
+
"special": false
|
19 |
+
}
|
20 |
+
},
|
21 |
+
"additional_special_tokens": [],
|
22 |
+
"bos_token": "<|startoftext|>",
|
23 |
+
"clean_up_tokenization_spaces": true,
|
24 |
+
"do_lower_case": true,
|
25 |
+
"eos_token": "<|endoftext|>",
|
26 |
+
"errors": "replace",
|
27 |
+
"model_max_length": 77,
|
28 |
+
"pad_token": "<|endoftext|>",
|
29 |
+
"processor_class": "CLIPProcessor",
|
30 |
+
"tokenizer_class": "CLIPTokenizer",
|
31 |
+
"tokenizer_file": "/Users/georgebredis/.cache/huggingface/hub/models--openai--clip-vit-base-patch32/snapshots/e6a30b603a447e251fdaca1c3056b2a16cdfebeb/tokenizer.json",
|
32 |
+
"unk_token": "<|endoftext|>"
|
33 |
+
}
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|