visheratin
commited on
Commit
•
35f7174
1
Parent(s):
4a0f492
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: cc-by-nc-4.0
|
3 |
+
datasets:
|
4 |
+
- visheratin/laion-coco-nllb
|
5 |
+
---
|
6 |
+
|
7 |
+
The code to run the model:
|
8 |
+
|
9 |
+
```
|
10 |
+
from transformers import AutoTokenizer, CLIPProcessor
|
11 |
+
import requests
|
12 |
+
from PIL import Image
|
13 |
+
|
14 |
+
from modeling_nllb_clip import NLLBCLIPModel # local file from the repo
|
15 |
+
|
16 |
+
processor = CLIPProcessor.from_pretrained("openai/clip-vit-base-patch32")
|
17 |
+
processor = processor.image_processor
|
18 |
+
tokenizer = AutoTokenizer.from_pretrained(
|
19 |
+
"facebook/nllb-200-distilled-600M"
|
20 |
+
)
|
21 |
+
image_path = "https://huggingface.co/spaces/jjourney1125/swin2sr/resolve/main/samples/butterfly.jpg"
|
22 |
+
image = Image.open(requests.get(image_path, stream=True).raw)
|
23 |
+
image_inputs = processor(images=image, return_tensors="pt")
|
24 |
+
text_inputs = tokenizer(
|
25 |
+
["cat", "dog", "butterfly"],
|
26 |
+
padding="longest",
|
27 |
+
return_tensors="pt",
|
28 |
+
)
|
29 |
+
|
30 |
+
hf_model = NLLBCLIPModel.from_pretrained("visheratin/nllb-clip-base")
|
31 |
+
|
32 |
+
outputs = hf_model(input_ids = text_inputs.input_ids, attention_mask = text_inputs.attention_mask, pixel_values=image_inputs.pixel_values)
|
33 |
+
```
|