Update app.py

app.py

@@ -881,7 +881,7 @@ num_inference_steps = 50  #@param           # Number of denoising steps
 guidance_scale = 8 #@param               # Scale for classifier-free guidance
 generator = torch.manual_seed(0)   # Seed generator to create the inital latent noise
 batch_size = 1
-blue_loss_scale = 200 #@param
+orange_loss_scale = 200 #@param
 
 # Prep text
 text_input = tokenizer([prompt], padding="max_length", max_length=tokenizer.model_max_length, truncation=True, return_tensors="pt")
@@ -936,7 +936,7 @@ for i, t in tqdm(enumerate(scheduler.timesteps), total=len(scheduler.timesteps))
         denoised_images = vae.decode((1 / 0.18215) * latents_x0).sample / 2 + 0.5 # range (0, 1)
 
         # Calculate loss
-        loss = orange_loss(denoised_images) * blue_loss_scale
+        loss = orange_loss(denoised_images) * orange_loss_scale
 
         # Occasionally print it out
         if i%10==0:
@@ -963,7 +963,7 @@ num_inference_steps = 50  #@param           # Number of denoising steps
 guidance_scale = 8 #@param               # Scale for classifier-free guidance
 generator = torch.manual_seed(77)   # Seed generator to create the inital latent noise
 batch_size = 1
-blue_loss_scale = 200 #@param
+orange_loss_scale = 200 #@param
 
 # Prep text
 text_input = tokenizer([prompt], padding="max_length", max_length=tokenizer.model_max_length, truncation=True, return_tensors="pt")
@@ -1018,7 +1018,7 @@ for i, t in tqdm(enumerate(scheduler.timesteps), total=len(scheduler.timesteps))
         denoised_images = vae.decode((1 / 0.18215) * latents_x0).sample / 2 + 0.5 # range (0, 1)
 
         # Calculate loss
-        loss = orange_loss(denoised_images) * blue_loss_scale
+        loss = orange_loss(denoised_images) * orange_loss_scale
 
         # Occasionally print it out
         if i%10==0:
@@ -1045,7 +1045,7 @@ num_inference_steps = 50  #@param           # Number of denoising steps
 guidance_scale = 8 #@param               # Scale for classifier-free guidance
 generator = torch.manual_seed(42)   # Seed generator to create the inital latent noise
 batch_size = 1
-blue_loss_scale = 200 #@param
+orange_loss_scale = 200 #@param
 
 # Prep text
 text_input = tokenizer([prompt], padding="max_length", max_length=tokenizer.model_max_length, truncation=True, return_tensors="pt")
@@ -1100,7 +1100,7 @@ for i, t in tqdm(enumerate(scheduler.timesteps), total=len(scheduler.timesteps))
         denoised_images = vae.decode((1 / 0.18215) * latents_x0).sample / 2 + 0.5 # range (0, 1)
 
         # Calculate loss
-        loss = orange_loss(denoised_images) * blue_loss_scale
+        loss = orange_loss(denoised_images) * orange_loss_scale
 
         # Occasionally print it out
         if i%10==0: