Update app.py

app.py

@@ -1,12 +1,16 @@
 import torch
 import torch.nn as nn
 from torchvision import transforms, models
+from torchvision.models import DenseNet121_Weights
 from PIL import Image
 import gradio as gr
 
 # Load the pre-trained DenseNet-121 model
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-model = models.densenet121(pretrained=True)
+
+# Use DenseNet121 with updated weights parameter
+weights = DenseNet121_Weights.IMAGENET1K_V1
+model = models.densenet121(weights=weights)
 
 # Modify the classifier layer to output probabilities for 14 classes (pathologies)
 num_classes = 14
@@ -17,8 +21,8 @@ model.classifier = nn.Sequential(
 
 try:
     model.load_state_dict(torch.load('chexnet.pth', map_location=device))
-except Exception as e:
-    print(f"Error loading pre-trained weights: {e}")
+except FileNotFoundError:
+    print("Error loading pre-trained weights: 'chexnet.pth' file not found.")
 model.to(device)
 model.eval()
 
@@ -76,14 +80,13 @@ references = """
 4. Abid, A., et al. (2019). Gradio: Hassle-Free Sharing and Testing of Machine Learning Models. arXiv preprint arXiv:1906.02569.
 """
 
-# Gradio Interface
+# Gradio Interface without using deprecated parameters
 interface = gr.Interface(
     fn=predict_disease,
     inputs=gr.components.Image(type='pil'),  # Updated input component
-    outputs="label",  # Output is a dictionary of labels with probabilities
+    outputs=[gr.components.Label(label="Disease Probabilities"), gr.components.Textbox(label="References", value=references, lines=10)],
     title="CheXNet Pneumonia Detection",
     description="Upload a chest X-ray to detect the probability of 14 different diseases.",
-    additional_outputs=[gr.components.Textbox(label="References", value=references, lines=10)],  # Display references
 )
 
 # Launch the Gradio app