Update app.py

app.py

@@ -2,20 +2,20 @@
 
 import gradio as gr
 import torch
-from PIL import Image, ImageDraw, ImageFont
+from PIL import Image
 from model import load_model
 from utils import preprocess_image, decode_predictions
 import os
 
 # Load the model (ensure the path is correct)
 MODEL_PATH = "finetuned_recog_model.pth"
-FONT_PATH = "NotoSansEthiopic-Regular.ttf"  # Update the path to your font
+FONT_PATH = "NotoSansEthiopic-Regular.ttf"  # Path to your font
 
 # Check if model file exists
 if not os.path.exists(MODEL_PATH):
     raise FileNotFoundError(f"Model file not found at {MODEL_PATH}. Please provide the correct path.")
 
-# Check if font file exists
+# Check if font file exists (if you plan to use it for any visualization)
 if not os.path.exists(FONT_PATH):
     raise FileNotFoundError(f"Font file not found at {FONT_PATH}. Please provide the correct path.")
 
@@ -23,17 +23,13 @@ if not os.path.exists(FONT_PATH):
 device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
 model = load_model(MODEL_PATH, device=device)
 
-# Load the font for rendering Amharic text
-from matplotlib import font_manager as fm
-import matplotlib.pyplot as plt
-
-ethiopic_font = fm.FontProperties(fname=FONT_PATH, size=15)
-pil_font = ImageFont.truetype(FONT_PATH, size=20)
-
 def recognize_text(image: Image.Image) -> str:
     """
     Function to recognize text from an image.
     """
+    if image is None:
+        return "No image provided."
+    
     # Preprocess the image
     input_tensor = preprocess_image(image).unsqueeze(0).to(device)  # [1, 3, 224, 224]
     
@@ -44,29 +40,16 @@ def recognize_text(image: Image.Image) -> str:
     # Decode predictions
     recognized_texts = decode_predictions(log_probs)
     
+    # Assuming batch size of 1
     return recognized_texts[0]
 
-def recognize_and_overlay(image: Image.Image) -> Image.Image:
-    """
-    Function to recognize text and overlay it on the image.
-    """
-    recognized_text = recognize_text(image)
-    
-    # Overlay text on the image
-    draw = ImageDraw.Draw(image)
-    text_position = (10, 10)  # Top-left corner
-    text_color = (255, 0, 0)  # Red color
-    draw.text(text_position, f"Recognized: {recognized_text}", font=pil_font, fill=text_color)
-    
-    return image
-
 # Define Gradio Interface
 iface = gr.Interface(
-    fn=recognize_and_overlay,
+    fn=recognize_text,
     inputs=gr.Image(type="pil", label="Upload Image"),
-    outputs=gr.Image(type="pil", label="Image with Recognized Text"),
+    outputs=gr.Textbox(label="Recognized Amharic Text"),
     title="Amharic Text Recognition",
-    description="Upload an image containing Amharic text. The app will recognize and overlay the text on the image."
+    description="Upload an image containing Amharic text, and the model will recognize and display the text."
 )
 
 # Launch the Gradio app