Update app.py

app.py

@@ -22,14 +22,14 @@ split_dataset = dataset['train'].train_test_split(test_size=0.2, seed=42)  # 80%
 dataset['train'] = split_dataset['train']
 dataset['test'] = split_dataset['test']
 
-
 def create_url_from_title(title: str, timestamp: int):
     video_urls = load_dataset("eybro/video_urls")
     df = video_urls['train'].to_pandas()
+    print(df.to_string())
     filtered = df[df['title'] == title]
-    base_url = df["url"][0]
-    return base_url + f"?t={timestamp}s"
-
+    print(filtered)
+    base_url = filtered.iloc[0, :]["url"]
+    return base_url + f"&t={timestamp}s"
 
 def find_nearest_neighbors(encoded_images, input_image, top_n=5):
     """
@@ -88,8 +88,10 @@ def inference(image):
       print(im["label"], im["timestamp"])
 
     result_image = get_image(top4[0])
-    result = f"{result_image['label']} {result_image['timestamp']} \n{create_url_from_title(result_image['label'], result_image['timestamp'])}"
-
+    url = create_url_from_title(result_image['label'], result_image['timestamp'])
+    result = f"{result_image['label']} {result_image['timestamp']} \n{url}"
+    #result =  f"[This is a link to the video]({url})"
+    
     n=2
     plt.figure(figsize=(8, 8))
     for i, (image1, image2) in enumerate(zip(top4[:2], top4[2:])):
@@ -110,9 +112,17 @@ def inference(image):
     
     return result
     
-
-
-demo = gr.Interface(fn=inference, 
-                    inputs=gr.Image(label='Upload image'),
-                    outputs="text")
-demo.launch()
+with gr.Blocks() as demo:
+    gr.Markdown(
+    """
+    # Image to Video App
+    Upload an image from a Gordon Ramsay show to find the corresponding youtube video
+    """)
+    with gr.Column(scale=1):
+        inp = gr.Image(label='Upload image')
+    with gr.Column(scale=1):
+        out = gr.Markdown()
+    inp.change(inference, inp, out)
+
+if __name__ == "__main__":
+    demo.launch()