Initial README (example, correct license...)
Browse files
README.md
CHANGED
@@ -1,3 +1,35 @@
|
|
1 |
---
|
2 |
-
license:
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
license: cc-by-nc-4.0
|
3 |
---
|
4 |
+
|
5 |
+
# Conditional ViT - B/16 - Text
|
6 |
+
|
7 |
+
*Introduced in **Weakly-Supervised Conditional Embedding for Referred Visual Search**, Lepage et al. 2023*
|
8 |
+
|
9 |
+
[`Paper`](https://arxiv.org/abs/2306.02928) | [`Training Data`](https://huggingface.co/datasets/Slep/LAION-RVS-Fashion) | [`Training Code`](https://github.com/Simon-Lepage/CondViT-LRVSF) | [`Demo`](https://huggingface.co/spaces/Slep/CondViT-LRVSF-Demo)
|
10 |
+
|
11 |
+
## General Infos
|
12 |
+
|
13 |
+
Model finetuned from CLIP ViT-B/16 on LRVSF at 224x224. The conditioning text is preprocessed by a frozen [Sentence T5-XL](https://huggingface.co/sentence-transformers/sentence-t5-xl).
|
14 |
+
|
15 |
+
Research use only.
|
16 |
+
|
17 |
+
## How to Use
|
18 |
+
|
19 |
+
```python
|
20 |
+
from PIL import Image
|
21 |
+
import requests
|
22 |
+
from transformers import AutoProcessor, AutoModel
|
23 |
+
import torch
|
24 |
+
|
25 |
+
model = AutoModel.from_pretrained("Slep/CondViT-B16-txt")
|
26 |
+
processor = AutoProcessor.from_pretrained("Slep/CondViT-B16-txt")
|
27 |
+
|
28 |
+
url = "https://huggingface.co/datasets/Slep/LAION-RVS-Fashion/resolve/main/assets/108856.0.jpg"
|
29 |
+
img = Image.open(requests.get(url, stream=True).raw)
|
30 |
+
txt = "a brown bag"
|
31 |
+
|
32 |
+
inputs = processor(images=[img], texts=[txt])
|
33 |
+
raw_embedding = model(**inputs)
|
34 |
+
normalized_embedding = torch.nn.functional.normalize(raw_embedding, dim=-1)
|
35 |
+
```
|