Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -26,11 +26,27 @@ We show that a lot of this bulk is actually _unnecessary_. Instead of manually a
|
|
| 26 |
|
| 27 |
## Intended uses & limitations
|
| 28 |
|
| 29 |
-
Hiera can be used for image classification, feature extraction or masked image modeling.
|
| 30 |
|
| 31 |
### How to use
|
| 32 |
|
| 33 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
|
| 35 |
|
| 36 |
### BibTeX entry and citation info
|
|
|
|
| 26 |
|
| 27 |
## Intended uses & limitations
|
| 28 |
|
| 29 |
+
Hiera can be used for image classification, feature extraction or masked image modeling. This checkpoint in specific is intended for **Masked Image Modeling**.
|
| 30 |
|
| 31 |
### How to use
|
| 32 |
|
| 33 |
+
```python
|
| 34 |
+
from transformers import AutoImageProcessor, HieraForPreTraining
|
| 35 |
+
import torch
|
| 36 |
+
from PIL import Image
|
| 37 |
+
import requests
|
| 38 |
+
|
| 39 |
+
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
|
| 40 |
+
image = Image.open(requests.get(url, stream=True).raw)
|
| 41 |
+
|
| 42 |
+
image_processor = AutoImageProcessor.from_pretrained("facebook/hiera-tiny-224-mae-hf")
|
| 43 |
+
model = HieraForPreTraining.from_pretrained("facebook/hiera-tiny-224-mae-hf")
|
| 44 |
+
|
| 45 |
+
inputs = image_processor(images=image, return_tensors="pt")
|
| 46 |
+
outputs = model(**inputs)
|
| 47 |
+
logits = outputs.logits
|
| 48 |
+
loss = outputs.loss
|
| 49 |
+
```
|
| 50 |
|
| 51 |
|
| 52 |
### BibTeX entry and citation info
|