Update app.py

app.py

@@ -4,22 +4,50 @@ from generate_prompts import generate_prompt
 from diffusers import AutoPipelineForText2Image
 from io import BytesIO
 import gradio as gr
+import threading
 
-# Asynchronously load the model once outside of the function
-model = None
+# Load the model once outside of the function
+print("Loading the model...")
+model = AutoPipelineForText2Image.from_pretrained("stabilityai/sdxl-turbo")
+print("Model loaded successfully.")
 
-async def load_model():
-    global model
-    print("Loading the model...")
-    model = AutoPipelineForText2Image.from_pretrained("stabilityai/sdxl-turbo")
-    print("Model loaded successfully.")
+# Create a thread-local storage object
+thread_local = threading.local()
 
-asyncio.run(load_model())
+class Scheduler:
+    def __init__(self):
+        self._step = threading.local()
+        self._init_step_index()
 
-async def generate_image(prompt, prompt_name):
+    def _init_step_index(self):
+        self._step.step = 0
+        print(f"Initialized step index: {self._step.step}")
+
+    @property
+    def step(self):
+        return self._step.step
+
+    def step_process(self):
+        self._step.step += 1
+        print(f"Step index updated to: {self._step.step}")
+
+scheduler = Scheduler()
+
+def generate_image(prompt, prompt_name):
     try:
+        # Initialize step index for the current thread if not already done
+        if not hasattr(scheduler._step, 'step'):
+            scheduler._init_step_index()
+
+        print(f"Initial step index for {prompt_name}: {scheduler.step}")
         print(f"Generating response for {prompt_name} with prompt: {prompt}")
+        
         output = model(prompt=prompt, num_inference_steps=1, guidance_scale=0.0)
+        
+        # Update and print step index
+        scheduler.step_process()
+        print(f"Updated step index for {prompt_name}: {scheduler.step}")
+
         print(f"Output for {prompt_name}: {output}")
 
         # Check if the model returned images
@@ -48,12 +76,13 @@ async def queue_api_calls(sentence_mapping, character_dict, selected_style):
     for paragraph_number, sentences in sentence_mapping.items():
         combined_sentence = " ".join(sentences)
         print(f"combined_sentence for paragraph {paragraph_number}: {combined_sentence}")
-        prompt = generate_prompt(combined_sentence, sentence_mapping, character_dict, selected_style)  # Correct prompt generation
+        prompt = generate_prompt(combined_sentence, character_dict, selected_style)  # Correct prompt generation
         prompts.append((paragraph_number, prompt))
         print(f"Generated prompt for paragraph {paragraph_number}: {prompt}")
 
     # Generate images for each prompt in parallel
-    tasks = [generate_image(prompt, f"Prompt {paragraph_number}") for paragraph_number, prompt in prompts]
+    loop = asyncio.get_running_loop()
+    tasks = [loop.run_in_executor(None, generate_image, prompt, f"Prompt {paragraph_number}") for paragraph_number, prompt in prompts]
     print("Tasks created for image generation.")
     responses = await asyncio.gather(*tasks)
     print("Responses received from image generation tasks.")
@@ -73,6 +102,9 @@ def process_prompt(sentence_mapping, character_dict, selected_style):
         asyncio.set_event_loop(loop)
     print("Event loop created.")
 
+    # Initialize thread-local variables
+    scheduler._init_step_index()
+
     # This sends the prompts to function that sets up the async calls. Once all the calls to the API complete, it returns a list of the gr.Textbox with value= set.
     cmpt_return = loop.run_until_complete(queue_api_calls(sentence_mapping, character_dict, selected_style))
     print(f"process_prompt completed with return value: {cmpt_return}")