Update app.py

app.py

@@ -2,6 +2,7 @@ import gradio as gr
 import torch
 torch.jit.script = lambda f: f
 import timm
+import time
 from huggingface_hub import hf_hub_download
 from safetensors.torch import load_file
 from share_btn import community_icon_html, loading_icon_html, share_js
@@ -124,7 +125,6 @@ pipe.set_ip_adapter_scale(0.8)
 zoe = ZoeDetector.from_pretrained("lllyasviel/Annotators")
 zoe.to(device)
 
-original_pipe = copy.deepcopy(pipe)
 pipe.to(device)
 
 last_lora = ""
@@ -209,10 +209,18 @@ def generate_image(prompt, negative, face_emb, face_image, image_strength, image
     global last_fused
     if last_lora != repo_name:
         if(last_fused):
+            st = time.time()
             pipe.unfuse_lora()
             pipe.unload_lora_weights()
+            et = time.time()
+            elapsed_time = et - st
+            print('Unfuse and unload took: ', elapsed_time, 'seconds')
+        st = time.time()
         pipe.load_lora_weights(loaded_state_dict)
         pipe.fuse_lora(lora_scale)
+        et = time.time()
+        elapsed_time = et - st
+        print('Fuse and load took: ', elapsed_time, 'seconds')
         last_fused = True
         is_pivotal = sdxl_loras[selected_state_index]["is_pivotal"]
         if(is_pivotal):
@@ -220,7 +228,6 @@ def generate_image(prompt, negative, face_emb, face_image, image_strength, image
             text_embedding_name = sdxl_loras[selected_state_index]["text_embedding_weights"]
             embedding_path = hf_hub_download(repo_id=repo_name, filename=text_embedding_name, repo_type="model")
             state_dict_embedding = load_file(embedding_path)
-            print(state_dict_embedding)
             try:
                 pipe.unload_textual_inversion()
                 pipe.load_textual_inversion(state_dict_embedding["clip_l"], token=["<s0>", "<s1>"], text_encoder=pipe.text_encoder, tokenizer=pipe.tokenizer)