Update app.py

app.py

@@ -104,6 +104,7 @@ def apply_style(style_name: str, positive: str, negative: str = "") -> Tuple[str
     if not negative:
         negative = ""
     return p.replace("{prompt}", positive), n + negative
+    
 def load_and_prepare_model(model_id):
     model_dtypes = {"ford442/RealVisXL_V5.0_BF16": torch.bfloat16,}
     dtype = model_dtypes.get(model_id, torch.bfloat16)  # Default to float32 if not found
@@ -210,25 +211,22 @@ def randomize_seed_fn(seed: int, randomize_seed: bool) -> int:
         seed = random.randint(0, MAX_SEED)
     return seed
 
-def uploadNote():
-          # write note txt
-    filename= f'tst_B_{seed}.txt'
-    timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
+def uploadNote(prompt,num_inference_steps,guidance_scale,timestamp):
+    filename= f'tst_A_{timestamp}.txt'
     with open(filename, "w") as f:
-        f.write(f"Realvis 5.0 (Tester B): {seed} png\n")
+        f.write(f"Realvis 5.0 (Tester A) \n")
         f.write(f"Date/time: {timestamp} \n")
         f.write(f"Prompt: {prompt} \n")
         f.write(f"Steps: {num_inference_steps} \n")
         f.write(f"Guidance Scale: {guidance_scale} \n")
         f.write(f"SPACE SETUP: \n")
         f.write(f"Use Model Dtype: no \n")
-        f.write(f"Model Scheduler: Euler_a all_custom before cuda woth trailing \n")
-        f.write(f"Model VAE: sdxl-vae  to bfloat with cuda then attn_proc \n")
-        f.write(f"Model UNET: default  to bfloat with cuda  \n")
+        f.write(f"Model Scheduler: Euler_a all_custom before cuda \n")
+        f.write(f"Model VAE: sdxl-vae to bfloat safetensor=false before cuda then attn_proc / scale factor 8 \n")
+        f.write(f"Model UNET: default ford442/RealVisXL_V5.0_BF16 \n")
         f.write(f"Model HiDiffusion OFF \n")
-        f.write(f"Model do_resize OFF \n")
-        f.write(f"Model vae_scale_factor 8 \n")
-        f.write(f"now added packages.txt \n")
+        f.write(f"Model do_resize ON \n")
+        f.write(f"added torch to prereq and changed accellerate  \n")
     upload_to_ftp(filename) 
     
 @spaces.GPU(duration=30)
@@ -271,7 +269,8 @@ def generate_30(
         options["use_resolution_binning"] = True
     images = []
     pipe.scheduler.set_timesteps(num_inference_steps,device)
-    uploadNote()
+    timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
+    uploadNote(prompt,num_inference_steps,guidance_scale,timestamp)
     for i in range(0, num_images, BATCH_SIZE):
         batch_options = options.copy()
         batch_options["prompt"] = options["prompt"][i:i+BATCH_SIZE]
@@ -326,7 +325,8 @@ def generate_60(
         options["use_resolution_binning"] = True
     images = []
     pipe.scheduler.set_timesteps(num_inference_steps,device)
-    uploadNote()
+    timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
+    uploadNote(prompt,num_inference_steps,guidance_scale,timestamp)
     for i in range(0, num_images, BATCH_SIZE):
         batch_options = options.copy()
         batch_options["prompt"] = options["prompt"][i:i+BATCH_SIZE]
@@ -381,7 +381,8 @@ def generate_90(
         options["use_resolution_binning"] = True
     images = []
     pipe.scheduler.set_timesteps(num_inference_steps,device)
-    uploadNote()
+    timestamp = datetime.datetime.now().strftime("%Y%m%d_%H%M%S")
+    uploadNote(prompt,num_inference_steps,guidance_scale,timestamp)
     for i in range(0, num_images, BATCH_SIZE):
         batch_options = options.copy()
         batch_options["prompt"] = options["prompt"][i:i+BATCH_SIZE]