Update app.py

app.py

@@ -198,7 +198,7 @@ device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
 print('Using device:', device)
 model = load_vqgan_model(args.vqgan_config, args.vqgan_checkpoint).to(device)
 perceptor = clip.load(args.clip_model, jit=False)[0].eval().requires_grad_(False).to(device)
-def inference(text, seed, step_size, max_iterations, width, height, init_image, init_weight, target_images):
+def inference(text, seed, step_size, max_iterations, width, height, init_image, init_weight, target_images, cutn, cut_pow):
     all_frames = []
     size=[width, height]
     texts = text
@@ -239,7 +239,7 @@ def inference(text, seed, step_size, max_iterations, width, height, init_image,
     # perceptor.visual.positional_embedding.data=clamp_with_grad(clock,0,1)
     cut_size = perceptor.visual.input_resolution
     f = 2**(model.decoder.num_resolutions - 1)
-    make_cutouts = MakeCutouts(cut_size, args.cutn, cut_pow=args.cut_pow)
+    make_cutouts = MakeCutouts(cut_size, cutn, cut_pow=cut_pow)
     toksX, toksY = size[0] // f, size[1] // f
     sideX, sideY = toksX * f, toksY * f
     if args.vqgan_checkpoint == 'vqgan_openimages_f16_8192.ckpt':
@@ -378,7 +378,9 @@ gr.Interface(
     gr.inputs.Slider(minimum=200, maximum=600, default=256, label='height', step=1),
     gr.inputs.Image(type="file", label="Initial Image (Optional)", optional=True),
     gr.inputs.Slider(minimum=0.0, maximum=15.0, default=0.0, label='Initial Weight', step=1.0),
-    gr.inputs.Image(type="file", label="Target Image (Optional)", optional=True)
+    gr.inputs.Image(type="file", label="Target Image (Optional)", optional=True),
+    gr.inputs.Slider(minimum=1, maximum=4, default=1, label='cutn', step=1),
+    gr.inputs.Slider(minimum=1.0, maximum=4.0, default=1.0, label='cut_pow', step=1.0)
      ], 
     [gr.outputs.Image(type="numpy", label="Output Image"),gr.outputs.Video(label="Output Video")],
     title=title,