update README.md

README.md

@@ -10,4 +10,39 @@ pipeline_tag: image-to-text
 tags:
 - image
 - vision
----
+---
+
+Vision Encoder Decoder (ViT + GPT2) model that fine-tuned on [flickr8k-dataset](https://huggingface.co/datasets/atasoglu/flickr8k-dataset) for image-to-text task.
+
+Example:
+
+```py
+from transformers import VisionEncoderDecoderModel, ViTImageProcessor, AutoTokenizer
+import torch
+from PIL import Image
+
+# load models
+feature_extractor = ViTImageProcessor.from_pretrained("atasoglu/vit-gpt2-flickr8k")
+tokenizer = AutoTokenizer.from_pretrained("atasoglu/vit-gpt2-flickr8k")
+model = VisionEncoderDecoderModel.from_pretrained("atasoglu/vit-gpt2-flickr8k")
+
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model.to(device)
+
+# load image
+img = Image.open("example.jpg")
+
+# encode (extracting features)
+pixel_values = feature_extractor(images=[img], return_tensors="pt").pixel_values
+pixel_values = pixel_values.to(device)
+
+# generate caption
+output_ids = model.generate(pixel_values)
+
+# decode
+preds = tokenizer.batch_decode(output_ids, skip_special_tokens=True)
+
+print(preds)
+```
+
+For more, see [this](https://ankur3107.github.io/blogs/the-illustrated-image-captioning-using-transformers/) awesome blog.