Update app.py

app.py

@@ -1,7 +1,43 @@
-import gradio as gr
+from transformers import VisionEncoderDecoderModel, ViTImageProcessor, AutoTokenizer
+import torch
+from PIL import Image
 
+model = VisionEncoderDecoderModel.from_pretrained("nlpconnect/vit-gpt2-image-captioning")
+feature_extractor = ViTImageProcessor.from_pretrained("nlpconnect/vit-gpt2-image-captioning")
+tokenizer = AutoTokenizer.from_pretrained("nlpconnect/vit-gpt2-image-captioning")
 
-image_to_text = gr.Interface.load("models/nlpconnect/vit-gpt2-image-captioning")
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model.to(device)
 
-demo = gr.Interface(image_to_text, inputs="image", outputs="text")
-demo.launch()
+max_length = 16
+num_beams = 4
+gen_kwargs = {"max_length": max_length, "num_beams": num_beams}
+def image_to_text(image_paths):
+    images=[image_paths]
+    
+  pixel_values = feature_extractor(images=images, return_tensors="pt").pixel_values
+  pixel_values = pixel_values.to(device)
+
+  output_ids = model.generate(pixel_values, **gen_kwargs)
+
+  preds = tokenizer.batch_decode(output_ids, skip_special_tokens=True)
+  preds = [pred.strip() for pred in preds]
+  return preds[0]
+
+
+title = ""
+description = ""
+
+interface = gr.Interface(
+    fn=image_to_text, 
+    inputs=[
+        gr.inputs.Image(type="pil")
+    ],
+    outputs=gr.Image(), 
+    title=title,
+    description=description,
+    enable_queue=True
+
+)
+
+interface.launch(debug=True)