add model card for unicom-vit-b-16 (#1)
Browse files- add model card for unicom-vit-b-16 (8e9842b0cd6867b7f80952ef2606637ef4731ccc)
Co-authored-by: George <jmzzomg@users.noreply.huggingface.co>
README.md
ADDED
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
pipeline_tag: image-feature-extraction
|
4 |
+
---
|
5 |
+
|
6 |
+
ONNX port of [Unicom](https://arxiv.org/abs/2304.05884) model from [open-metric-learning](https://github.com/OML-Team/open-metric-learning).
|
7 |
+
|
8 |
+
This model is intended to be used for similarity search.
|
9 |
+
|
10 |
+
### Usage
|
11 |
+
|
12 |
+
Here's an example of performing inference using the model with [FastEmbed](https://github.com/qdrant/fastembed).
|
13 |
+
|
14 |
+
```py
|
15 |
+
from fastembed import ImageEmbedding
|
16 |
+
|
17 |
+
images = [
|
18 |
+
"./path/to/image1.jpg",
|
19 |
+
"./path/to/image2.jpg",
|
20 |
+
]
|
21 |
+
|
22 |
+
model = ImageEmbedding(model_name="Qdrant/Unicom-ViT-B-16")
|
23 |
+
embeddings = list(model.embed(images))
|
24 |
+
|
25 |
+
# [
|
26 |
+
# array([ 1.70463976e-02, -3.60863991e-02, 1.24569749e-02, -4.28437591e-02 , ...], dtype=float32),
|
27 |
+
# array([ 0.03675087, 0.00696867, -0.01495106, -0.02828627, ...], dtype=float32)
|
28 |
+
# ]
|
29 |
+
```
|