Update app.py

app.py

@@ -34,7 +34,7 @@ def get_lora_sd_pipeline(
     if base_model_name_or_path is None:
         raise ValueError("Please specify the base model name or path")
 
-    pipe = StableDiffusionPipeline.from_pretrained(base_model_name_or_path, torch_dtype=dtype).to(device)
+    pipe = StableDiffusionPipeline.from_pretrained(base_model_name_or_path, torch_dtype=dtype, safety_checker=None).to(device)
     pipe.unet = PeftModel.from_pretrained(pipe.unet, unet_sub_dir)
 
     if os.path.exists(text_encoder_sub_dir):
@@ -42,9 +42,9 @@ def get_lora_sd_pipeline(
             pipe.text_encoder, text_encoder_sub_dir
         )
 
-    if dtype in (torch.float16, torch.bfloat16):
-        pipe.unet.half()
-        pipe.text_encoder.half()
+    # if dtype in (torch.float16, torch.bfloat16):
+    #     pipe.unet.half()
+    #     pipe.text_encoder.half()
 
     pipe.to(device)
     return pipe
@@ -90,10 +90,8 @@ def infer(
     progress=gr.Progress(track_tqdm=True),
 ):  
     generator = torch.Generator(device).manual_seed(seed)
-    pipe = get_lora_sd_pipeline(base_model_name_or_path=model_id)    
-    pipe = pipe.to(device)
+    pipe = get_lora_sd_pipeline(base_model_name_or_path=model_id)   
     # pipe.fuse_lora(lora_scale=lora_scale)
-    pipe.safety_checker = None
     # prompt_embeds = encode_prompt(prompt, pipe.tokenizer, pipe.text_encoder)
     # negative_prompt_embeds = encode_prompt(negative_prompt, pipe.tokenizer, pipe.text_encoder)
     
@@ -157,12 +155,12 @@ with gr.Blocks(css=css, fill_height=True) as demo:
                 value=7.0,  # Replace with defaults that work for your model
             )
         with gr.Row():
-            lora_scale = gr.Slider(
-                label="LoRA scale",
-                minimum=0.0,
-                maximum=1.0,
-                step=0.1,
-                value=1.0,
+        #     lora_scale = gr.Slider(
+        #         label="LoRA scale",
+        #         minimum=0.0,
+        #         maximum=1.0,
+        #         step=0.1,
+        #         value=1.0,
             )
 
             num_inference_steps = gr.Slider(