Upload app.py

app.py

@@ -0,0 +1,38 @@
+from PIL import Image
+from transformers import VisionEncoderDecoderModel , ViTFeatureExtractor , PreTrainedTokenizerFast
+import gradio as gr
+
+model = VisionEncoderDecoderModel.from_pretrained("ydshieh/vit-gpt2-coco-en")
+vit_feature_extractor = ViTFeatureExtractor.from_pretrained("google/vit-base-patch32-224-in21k")
+tokenizer = PreTrainedTokenizerFast.from_pretrained("distilgpt2")
+
+
+def caption_images(image):
+  pixel_values = vit_feature_extractor(images=image,return_tensors="pt").pixel_values
+  encoder_outputs = model.generate(pixel_values.to('cpu'),num_beams=5)
+  generated_sentence = tokenizer.batch_decode(encoder_outputs,skip_special_tokens=True)
+
+  return (generated_sentence[0].strip())
+
+
+inputs = [
+    gr.components.Image(type='pil',label='Original Image')
+]
+
+outputs = [
+    gr.components.Textbox(label='Caption')
+]
+
+title = "Simple Image captioning Application"
+description = "Upload an image to see the caption generated"
+example =['/content/messi.jpg']
+
+gr.Interface(
+    caption_images,
+    inputs,
+    outputs,
+    title=title,
+    description = description,
+    examples = example,
+).launch(debug=True)
+