Upload folder using huggingface_hub

app.py

@@ -88,7 +88,8 @@ with gr.Blocks(css=css) as demo:
     with gr.Column(elem_id='container'):
         content_and_output = gr.Image(label='Content', show_label=False, type='pil', sources=['upload', 'webcam', 'clipboard'], format='jpg', show_download_button=False)
         style_dropdown = gr.Radio(choices=list(style_options.keys()), label='Style', info='Note: Adjustments automatically optimize for different styles.', value='Starry Night', type='value')
-        with gr.Accordion('Adjustments', open=False):
+        
+        with gr.Accordion('Adjustments', open=True):
             with gr.Group():
                 style_strength_slider = gr.Slider(label='Style Strength', minimum=1, maximum=100, step=1, value=50)
                 
@@ -138,12 +139,8 @@ with gr.Blocks(css=css) as demo:
         )
         
         examples = gr.Examples(
-            examples=[
-                ['./content_images/Bridge.jpg', 'Starry Night', *optimal_settings['Starry Night']],
-                ['./content_images/GoldenRetriever.jpg', 'Lego Bricks', *optimal_settings['Lego Bricks']],
-                ['./content_images/SeaTurtle.jpg', 'Oil Painting', *optimal_settings['Oil Painting']],
-                ['./content_images/NYCSkyline.jpg', 'Scream', *optimal_settings['Scream']]
-            ],
+            label='Example',
+            examples=[['./content_images/Bridge.jpg', 'Starry Night', 100, False]],
             inputs=[content_and_output, style_dropdown, style_strength_slider, output_quality]
         )
 

inference.py

@@ -26,23 +26,25 @@ def inference(
     content_image,
     style_features,
     lr,
-    iterations=35,
+    iterations=3,
+    optim_caller=optim.LBFGS,
     alpha=1,
     beta=1
 ):
     generated_image = content_image.clone().requires_grad_(True)
-    optimizer = optim.AdamW([generated_image], lr=lr)
+    optimizer = optim_caller([generated_image], lr=lr)
 
     with torch.no_grad():
         content_features = model(content_image)
-    
-    for _ in tqdm(range(iterations), desc='The magic is happening ✨'):
+        
+    def closure():
         optimizer.zero_grad()
-
         generated_features = model(generated_image)
         total_loss = _compute_loss(generated_features, content_features, style_features, alpha, beta)
-
         total_loss.backward()
-        optimizer.step()
+        return total_loss
+    
+    for _ in tqdm(range(iterations), desc='The magic is happening ✨'):
+        optimizer.step(closure)
     
     return generated_image