Update app.py

app.py

@@ -1,64 +1,50 @@
+from PIL import Image
+import torch
+from transformers import NougatProcessor, VisionEncoderDecoderModel
 import gradio as gr
-from huggingface_hub import InferenceClient
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+# Load model and processor once at startup
+processor = NougatProcessor.from_pretrained("MohamedRashad/arabic-small-nougat")
+model = VisionEncoderDecoderModel.from_pretrained("MohamedRashad/arabic-small-nougat")
 
+device = "cuda" if torch.cuda.is_available() else "cpu"
+model.to(device)
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+context_length = 2048
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+def predict(image):
+    # Ensure image is in RGB format
+    image = image.convert("RGB")
 
-    messages.append({"role": "user", "content": message})
+    # Prepare input
+    pixel_values = processor(images=image, return_tensors="pt").pixel_values
 
-    response = ""
+    # Generate transcription
+    outputs = model.generate(
+        pixel_values.to(device),
+        min_length=1,
+        max_new_tokens=context_length,
+        bad_words_ids=[[processor.tokenizer.unk_token_id]],
+    )
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+    # Decode output
+    page_sequence = processor.batch_decode(outputs, skip_special_tokens=True)[0]
+    page_sequence = processor.post_process_generation(page_sequence, fix_markdown=False)
 
-        response += token
-        yield response
+    return page_sequence
 
+# Gradio Interface
+title = "Arabic Nougat OCR - Handwritten & Printed Document Recognizer"
+description = "Transcribe Arabic documents using a fine-tuned Nougat model."
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
+interface = gr.Interface(
+    fn=predict,
+    inputs=gr.Image(type="pil", label="Upload an Arabic Document"),
+    outputs=gr.Textbox(label="Transcription", lines=15),
+    title=title,
+    description=description,
+    examples=[["example_1.jpg"], ["example_2.jpg"]]
 )
 
-
 if __name__ == "__main__":
-    demo.launch()
+    interface.launch()