first commit

app.py

@@ -0,0 +1,33 @@
+import gradio as gr
+from transformers import VisionEncoderDecoderModel, ViTImageProcessor, AutoTokenizer
+from PIL import Image
+import torch
+
+# Load ViT-GPT2 (Apache-2.0 licensed, safe to use)
+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")
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+model.to(device)
+
+def caption_image(image):
+    # Convert image to tensor
+    pixel_values = feature_extractor(images=image, return_tensors="pt").pixel_values.to(device)
+
+    # Generate caption
+    output_ids = model.generate(pixel_values, max_length=50, num_beams=4)
+    caption = tokenizer.decode(output_ids[0], skip_special_tokens=True)
+    return caption
+
+# Build Gradio app
+demo = gr.Interface(
+    fn=caption_image,
+    inputs=gr.Image(type="pil"),
+    outputs="text",
+    title="Chart Analyzer",
+    description="Upload a chart/visualization image and get a description of it."
+)
+
+if __name__ == "__main__":
+    demo.launch()