Update app.py

app.py

@@ -1,20 +1,23 @@
 import gradio as gr
 import torch
-from transformers import AutoModel, AutoTokenizer
+import requests
+from torchvision import transforms
 
-# Load the model and tokenizer
-model = AutoModel.from_pretrained("detectron2/faster_rcnn_R_50_FPN_3x")
-tokenizer = AutoTokenizer.from_pretrained("detectron2/faster_rcnn_R_50_FPN_3x")
+model = torch.hub.load('pytorch/vision:v0.6.0', 'resnet18', pretrained=True).eval()
+response = requests.get("https://git.io/JJkYN")
+labels = response.text.split("\n")
 
-# Define a function that takes an image as input and returns the output of the model
-def detect_faces(image):
-    inputs = {"image": image}
-    output = model(**inputs)
-    boxes = output[0]["boxes"]
-    return boxes
+def predict(inp):
+  inp = transforms.ToTensor()(inp).unsqueeze(0)
+  with torch.no_grad():
+    prediction = torch.nn.functional.softmax(model(inp)[0], dim=0)
+    confidences = {labels[i]: float(prediction[i]) for i in range(1000)}    
+  return confidences
 
-# Create the Gradio interface
-interface = gr.Interface(fn=detect_faces, inputs=gr.inputs.Image(), outputs=gr.outputs.Boxes())
-
-# Launch the interface
-interface.launch()
+demo = gr.Interface(fn=predict, 
+             inputs=gr.inputs.Image(type="pil"),
+             outputs=gr.outputs.Label(num_top_classes=3),
+             examples=[["cheetah.jpg"]],
+             )
+             
+demo.launch()