added auto filled sfts + trigger word

app.py

@@ -1,10 +1,14 @@
 import gradio as gr
-from huggingface_hub import login
+from huggingface_hub import login, HfFileSystem, HfApi, ModelCard
 import os 
+
 is_shared_ui = True if "fffiloni/sd-xl-custom-model" in os.environ['SPACE_ID'] else False
 hf_token = os.environ.get("HF_TOKEN")
 login(token=hf_token)
 
+fs = HfFileSystem(token=hf_token)
+api = HfApi()
+
 import torch
 from diffusers import DiffusionPipeline, AutoencoderKL
 
@@ -13,25 +17,64 @@ vae = AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype
 pipe = DiffusionPipeline.from_pretrained(
     "stabilityai/stable-diffusion-xl-base-1.0",
     vae=vae, torch_dtype=torch.float16, variant="fp16",
-    use_safetensors=True
+    #use_safetensors=True
 )
 
 device="cuda" if torch.cuda.is_available() else "cpu"
 
 pipe.to(device)
 
-def load_model(custom_model, weight_name):
-    
+def load_model(custom_model):
+
     if custom_model == "":
         gr.Warning("If you want to use a private model, you need to duplicate this space on your personal account.")
         raise gr.Error("You forgot to define Model ID.")
-    
-    # This is where you load your trained weights
-    pipe.load_lora_weights(custom_model, weight_name=weight_name, use_auth_token=True)
-    
-    return "Model loaded!"
 
-def infer (prompt, inf_steps, guidance_scale, seed, lora_weight, progress=gr.Progress(track_tqdm=True)):
+    # Get instance_prompt a.k.a trigger word
+    card = ModelCard.load(custom_model)
+    repo_data = card.data.to_dict()
+    instance_prompt = repo_data.get("instance_prompt")
+
+    if instance_prompt is not None:
+        print(f"Trigger word: {instance_prompt}")
+    else:
+        instance_prompt = "no trigger word needed"
+        print(f"Trigger word: no trigger word needed")
+
+    # List all ".safetensors" files in repo
+    sfts_available_files = fs.glob(f"{custom_model}/*safetensors")
+    sfts_available_files = get_files(sfts_available_files)
+
+    if sfts_available_files == []:
+        sfts_available_files = ["NO SAFETENSORS FILE"]
+
+    print(f"Safetensors available: {sfts_available_files}")
+
+    return gr.update(choices=sfts_available_files, value=sfts_available_files[0], visible=True), gr.update(value=instance_prompt, visible=True)
+
+
+
+def infer (custom_model, weight_name, prompt, inf_steps, guidance_scale, seed, lora_weight, progress=gr.Progress(track_tqdm=True)):
+
+    if weight_name == "NO SAFETENSORS FILE": 
+        pipe.load_lora_weights(
+            custom_model,     
+            low_cpu_mem_usage = True,
+            use_auth_token = True
+        )
+
+    else:
+        pipe.load_lora_weights(
+            custom_model,
+            weight_name = weight_name,        
+            low_cpu_mem_usage = True,
+            use_auth_token = True
+        )
+
+    pipe.fuse_lora(lora_weight)
+
+    if seed < 0 :
+        seed = random.randint(0, 423538377342)
     
     generator = torch.Generator(device="cuda").manual_seed(seed)
     
@@ -97,11 +140,34 @@ with gr.Blocks(css=css) as demo:
         """)
         with gr.Row():
             with gr.Column():
-                custom_model = gr.Textbox(label="Your custom model ID", placeholder="your_username/your_trained_model_name", info="Make sure your model is set to PUBLIC ")
-                weight_name = gr.Textbox(label="Safetensors file", value="pytorch_lora_weights.safetensors", info="specify which one if model has several .safetensors files")
+                if not is_shared_ui:
+                    your_username = api.whoami()["name"]
+                    my_models = api.list_models(author=your_username, filter=["diffusers", "stable-diffusion-xl", 'lora'])
+                    model_names = [item.modelId for item in my_models]
+
+                if not is_shared_ui:
+                    custom_model = gr.Dropdown(
+                        label = "Your custom model ID",
+                        choices = model_names,
+                        allow_custom_value = True
+                        #placeholder = "username/model_id"
+                    )
+                else:
+                    custom_model = gr.Textbox(
+                        label="Your custom model ID", 
+                        placeholder="your_username/your_trained_model_name", 
+                        info="Make sure your model is set to PUBLIC"
+                    )
+                
+                weight_name = gr.Dropdown(
+                    label="Safetensors file", 
+                    #value="pytorch_lora_weights.safetensors", 
+                    info="specify which one if model has several .safetensors files",
+                    visible = False
+                )
             with gr.Column():
                 load_model_btn = gr.Button("Load my model")
-                model_status = gr.Textbox(label="model status", interactive=False)
+                trigger_word = gr.Textbox(label="Trigger word", interactive=False, visible=False)
         
         prompt_in = gr.Textbox(label="Prompt")   
         with gr.Row():
@@ -121,10 +187,11 @@ with gr.Blocks(css=css) as demo:
             )
             seed = gr.Slider(
                 label="Seed",
-                minimum=0,
-                maximum=500000,
+                info = "-1 denotes a random seed",
+                minimum=-1,
+                maximum=423538377342,
                 step=1,
-                value=42
+                value=-1
             )
         lora_weight = gr.Slider(
             label="LoRa weigth",
@@ -138,12 +205,12 @@ with gr.Blocks(css=css) as demo:
 
     load_model_btn.click(
         fn = load_model,
-        inputs=[custom_model, weight_name],
-        outputs = [model_status]
+        inputs=[custom_model],
+        outputs = [model_status, weight_name, trigger_word]
     )
     submit_btn.click(
         fn = infer,
-        inputs = [prompt_in, inf_steps, guidance_scale, seed, lora_weight],
+        inputs = [custom_model, weight_name, prompt_in, inf_steps, guidance_scale, seed, lora_weight],
         outputs = [image_out]
     )