Update app.py

app.py

@@ -1,39 +1,66 @@
 import gradio as gr
-from transformers import TrOCRProcessor, VisionEncoderDecoderModel
-from PIL import Image
 import torch
-import numpy as np
+from PIL import Image
+from transformers import TrOCRProcessor, VisionEncoderDecoderModel
+import os
+import json
 
-# Load the pre-trained model and processor
+# Load the model and processor
 processor = TrOCRProcessor.from_pretrained('microsoft/trocr-base-handwritten')
 model = VisionEncoderDecoderModel.from_pretrained('microsoft/trocr-base-handwritten')
 
-# Define the prediction function
+# Load feedback data
+feedback_data_path = "feedback_data.json"
+if os.path.exists(feedback_data_path):
+    with open(feedback_data_path, "r") as file:
+        feedback_data = json.load(file)
+else:
+    feedback_data = []
+
 def recognize_handwriting(image):
-    image = Image.fromarray(image.astype('uint8'), 'RGB')
+    if isinstance(image, dict):
+        image = Image.fromarray(image['image'].astype('uint8')).convert("RGB")
+    else:
+        image = Image.open(image).convert("RGB")
+
     pixel_values = processor(images=image, return_tensors="pt").pixel_values
     generated_ids = model.generate(pixel_values)
     generated_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
+    
     return generated_text
 
-def provide_feedback(image, correct_text):
-    # Save the feedback to a file or database for later retraining
-    with open("feedback.txt", "a") as f:
-        f.write(f"{correct_text}\n")
+def provide_feedback(image, corrected_text):
+    feedback_data.append({
+        "image": image,
+        "corrected_text": corrected_text
+    })
+    
+    # Save feedback every 5 entries
+    if len(feedback_data) % 5 == 0:
+        with open(feedback_data_path, "w") as file:
+            json.dump(feedback_data, file)
+    
     return "Feedback received. Thank you!"
 
+# Gradio Interface
 with gr.Blocks() as demo:
-    with gr.Tab("Recognize Handwriting"):
-        sketchpad = gr.Sketchpad(label="Draw something")
-        output = gr.Textbox(label="Recognized Text")
-        recognize_button = gr.Button("Recognize")
-        recognize_button.click(fn=recognize_handwriting, inputs=sketchpad, outputs=output)
-
-    with gr.Tab("Provide Feedback"):
-        image_feedback = gr.Image(type="numpy")
-        correct_text = gr.Textbox(label="Correct Text")
-        feedback_button = gr.Button("Submit Feedback")
-        feedback_output = gr.Textbox()
-        feedback_button.click(fn=provide_feedback, inputs=[image_feedback, correct_text], outputs=feedback_output)
+    gr.Markdown("# Handwriting Recognition with Feedback")
+    
+    with gr.Tabs():
+        with gr.TabItem("Recognize Handwriting"):
+            image_input = gr.Image(source="upload", tool="sketch", type="numpy", label="Draw or Upload an Image")
+            recognize_button = gr.Button("Recognize Handwriting")
+            output_text = gr.Textbox(label="Recognized Text")
+            
+            recognize_button.click(recognize_handwriting, inputs=image_input, outputs=output_text)
+        
+        with gr.TabItem("Provide Feedback"):
+            feedback_image_input = gr.Image(source="upload", tool="sketch", type="numpy", label="Draw or Upload an Image")
+            corrected_text_input = gr.Textbox(label="Corrected Text")
+            feedback_button = gr.Button("Submit Feedback")
+            feedback_output = gr.Textbox(label="Feedback Status")
+            
+            feedback_button.click(provide_feedback, inputs=[feedback_image_input, corrected_text_input], outputs=feedback_output)
 
-demo.launch()
+# Run the Gradio app
+demo.launch(share=True)