Create app.py

app.py

@@ -0,0 +1,37 @@
+import gradio as gr
+from transformers import pipeline
+from PIL import Image
+
+# Load both models
+model_pipeline_v1 = pipeline(task="image-classification", model="ppicazo/autotrain-ap-pass-fail-v1")
+model_pipeline_v2 = pipeline(task="image-classification", model="ppicazo/allsky-stars-detected-v2")
+
+def predict(image):
+    # Resize the image to have width 1080 while keeping the aspect ratio
+    width = 1080
+    ratio = width / image.width
+    height = int(image.height * ratio)
+    resized_image = image.resize((width, height))
+    
+    # Perform predictions with both models
+    predictions_v1 = model_pipeline_v1(resized_image)
+    predictions_v2 = model_pipeline_v2(resized_image)
+    
+    # Format the results for each model
+    results_v1 = {p["label"]: p["score"] for p in predictions_v1}
+    results_v2 = {p["label"]: p["score"] for p in predictions_v2}
+    
+    # Return results as separate outputs
+    return results_v1, results_v2
+
+# Define the Gradio Interface
+gr.Interface(
+    fn=predict,
+    inputs=gr.Image(type="pil", label="Upload image"),
+    outputs=[
+        gr.Label(num_top_classes=5, label="Model v1 Predictions"),
+        gr.Label(num_top_classes=5, label="Model v2 Predictions"),
+    ],
+    title="Star Detector (Two Models)",
+    allow_flagging="manual",
+).launch()