nielsr HF staff commited on
Commit
212b600
1 Parent(s): fef36e9

Update README

Browse files
Files changed (1) hide show
  1. README.md +3 -2
README.md CHANGED
@@ -41,11 +41,12 @@ url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
41
  image = Image.open(requests.get(url, stream=True).raw)
42
  feature_extractor = ViTFeatureExtractor.from_pretrained('google/vit-base-patch32-384')
43
  model = ViTForImageClassification.from_pretrained('google/vit-base-patch32-384')
44
- inputs = feature_extractor(images=image)
45
  outputs = model(**inputs)
46
  logits = outputs.logits
47
  # model predicts one of the 1000 ImageNet classes
48
- predicted_class = logits.argmax(-1).item()
 
49
  ```
50
 
51
  Currently, both the feature extractor and model support PyTorch. Tensorflow and JAX/FLAX are coming soon, and the API of ViTFeatureExtractor might change.
 
41
  image = Image.open(requests.get(url, stream=True).raw)
42
  feature_extractor = ViTFeatureExtractor.from_pretrained('google/vit-base-patch32-384')
43
  model = ViTForImageClassification.from_pretrained('google/vit-base-patch32-384')
44
+ inputs = feature_extractor(images=image, return_tensors="pt")
45
  outputs = model(**inputs)
46
  logits = outputs.logits
47
  # model predicts one of the 1000 ImageNet classes
48
+ predicted_class_idx = logits.argmax(-1).item()
49
+ print("Predicted class:", model.config.id2label[predicted_class_idx])
50
  ```
51
 
52
  Currently, both the feature extractor and model support PyTorch. Tensorflow and JAX/FLAX are coming soon, and the API of ViTFeatureExtractor might change.