Update app.py

app.py

@@ -164,10 +164,13 @@ def run_lora(prompt, negative, lora_scale, selected_state, sdxl_loras, sdxl_lora
     
     full_path_lora = state_dicts[repo_name]["saved_name"]
     loaded_state_dict = copy.deepcopy(state_dicts[repo_name]["state_dict"])
-    cross_attention_kwargs = None
+    #cross_attention_kwargs = None
     if last_lora != repo_name:
+        #if(last_fused):
+            #pipe.unfuse_lora()
         pipe.load_lora_weights(loaded_state_dict, adapter_name=sdxl_loras[selected_state.index]["repo"])
-        pipe.set_adapters([sdxl_loras[selected_state.index]["repo"], "lcm_lora"], adapter_weights=[0.8, 1.0])
+        pipe.set_adapters([sdxl_loras[selected_state.index]["repo"], "lcm_lora"], adapter_weights=[lora_scale, 1.0])
+        #last_fused = True
         is_pivotal = sdxl_loras[selected_state.index]["is_pivotal"]
         if(is_pivotal):
             #Add the textual inversion embeddings from pivotal tuning models
@@ -182,7 +185,8 @@ def run_lora(prompt, negative, lora_scale, selected_state, sdxl_loras, sdxl_lora
         prompt=prompt,
         negative_prompt=negative,
         num_inference_steps=4,
-        guidance_scale=0.5,
+        guidance_scale=0.5
+        
     ).images[0]
     
     last_lora = repo_name
@@ -207,7 +211,8 @@ with gr.Blocks(css="custom.css") as demo:
     gr_sdxl_loras = gr.State(value=sdxl_loras_raw)
     gr_sdxl_loras_new = gr.State(value=sdxl_loras_raw_new)
     title = gr.HTML(
-        """<h1><img src="https://i.imgur.com/vT48NAO.png" alt="LoRA">LCM LoRA the Explorer</h1>
+        """<h1><img src="https://i.imgur.com/vT48NAO.png" alt="LoRA"><i>Experimental</i> LCM LoRA the Explorer</h1>
+        <br>
         Combine loading an [LCM LoRA](#) with your favorite SDXL LoRa and run LoRAs in only 4 steps. Check out our blog to see how this works.
         """,
         elem_id="title",