Update app.py

app.py

@@ -1,78 +1,87 @@
-from transformers import AutoConfig, AutoTokenizer, AutoModelForSeq2SeqLM, AutoModelForCausalLM, MistralForCausalLM
+from transformers import AutoConfig, AutoTokenizer, AutoModelForCausalLM
 from peft import PeftModel, PeftConfig
 import torch
 import gradio as gr
-import random
-from textwrap import wrap
+import json
+import os
+import shutil
+import requests
 
-EXAMPLES = [
-    ["Hey Falcon! Any recommendations for my holidays in Abu Dhabi?"],
-    ["What's the Everett interpretation of quantum mechanics?"],
-    ["Give me a list of the top 10 dive sites you would recommend around the world."],
-    ["Can you tell me more about deep-water soloing?"],
-    ["Can you write a short tweet about the release of our latest AI model, Falcon LLM?"]
-    ]
+# Define the device
+device = "cuda" if torch.cuda.is_available() else "cpu"
 
 
-device = "cuda" if torch.cuda.is_available() else "cpu"
+# Use model IDs as variables
 base_model_id = "tiiuae/falcon-7b-instruct"
 model_directory = "Tonic/GaiaMiniMed"
 
+# Instantiate the Tokenizer
 tokenizer = AutoTokenizer.from_pretrained(base_model_id, trust_remote_code=True, padding_side="left")
+tokenizer.pad_token = tokenizer.eos_token
+tokenizer.padding_side = 'left'
+
+
+# Load the GaiaMiniMed model with the specified configuration
+# Load the Peft model with a specific configuration
+# Specify the configuration class for the model
 model_config = AutoConfig.from_pretrained(base_model_id)
+# Load the PEFT model with the specified configuration
 peft_model = AutoModelForCausalLM.from_pretrained(model_directory, config=model_config)
 peft_model = PeftModel.from_pretrained(peft_model, model_directory)
 
-def format_prompt(message, history, system_prompt):
-  prompt = ""
-  if system_prompt:
-    prompt += f"System: {system_prompt}\n"
-  for user_prompt, bot_response in history:
-    prompt += f"User: {user_prompt}\n"
-    prompt += f"Falcon: {bot_response}\n" # Response already contains "Falcon: "
-  prompt += f"""User: {message}
-Falcon:"""
-  return prompt
-
-seed = 42
-
-def generate(
-    prompt, history, system_prompt="", temperature=0.9, max_new_tokens=500, top_p=0.95, repetition_penalty=1.0,
-):
-    temperature = float(temperature)
-    if temperature < 1e-2:
-        temperature = 1e-2
-    top_p = float(top_p)
-    global seed
-    generate_kwargs = dict(
-        temperature=temperature,
-        max_new_tokens=max_new_tokens,
-        top_p=top_p,
-        repetition_penalty=1.0,
-        stop_sequences="[END]",
-        do_sample=True,
-        seed=seed,
-    )
-    seed = seed + 1
-    formatted_prompt = format_prompt(prompt, history, system_prompt)
 
-    try:
-        stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
-        output = ""
 
-        for response in stream:
-            output += response.token.text
+# Class to encapsulate the Falcon chatbot
+class FalconChatBot:
+    def __init__(self, system_prompt="You are an expert medical analyst:"):
+        self.system_prompt = system_prompt
+
+    def process_history(self, history):
+        # Filter out special commands from the history
+        filtered_history = []
+        for message in history:
+            user_message = message["user"]
+            assistant_message = message["assistant"]
+            # Check if the user_message is not a special command
+            if not user_message.startswith("Falcon:"):
+                filtered_history.append({"user": user_message, "assistant": assistant_message})
+        return filtered_history
+
+    def predict(self, system_prompt, user_message, assistant_message, history, max_length=500):
+        # Process the history to remove special commands
+        processed_history = self.process_history(history)
+
+        # Combine the user and assistant messages into a conversation
+        conversation = f"{system_prompt}\nFalcon: {assistant_message if assistant_message else ''} User: {user_message}\nFalcon:\n"
     
-            for stop_str in STOP_SEQUENCES:
-                if output.endswith(stop_str):
-                    output = output[:-len(stop_str)]
-                    output = output.rstrip()
-                    yield output
-            yield output
-    except Exception as e:
-        raise gr.Error(f"Error while generating: {e}")
-    return output
+        # Encode the conversation using the tokenizer
+        input_ids = tokenizer.encode(conversation, return_tensors="pt", add_special_tokens=False)
+
+        # Generate a response using the Falcon model
+        response_text = peft_model.generate(input_ids, max_length=max_length, use_cache=True, early_stopping=True, bos_token_id=peft_model.config.bos_token_id, eos_token_id=peft_model.config.eos_token_id, pad_token_id=peft_model.config.eos_token_id, temperature=0.4, do_sample=True)
 
+        # Generate the formatted conversation in Falcon message format
+        conversation = f"{system_prompt}\n"
+        for message in processed_history:
+            user_message = message["user"]
+            assistant_message = message["assistant"]
+            conversation += f"Falcon:{' ' + assistant_message if assistant_message else ''} User: {user_message}\n Falcon:\n"
+
+        return response_text
+
+
+
+# Create the Falcon chatbot instance
+falcon_bot = FalconChatBot()
+
+# Define the Gradio interface
+title = "👋🏻Welcome to Tonic's 🦅Falcon's Medical👨🏻‍⚕️Expert Chat🚀"
+description = "You can use this Space to test out the GaiaMiniMed model [(Tonic/GaiaMiniMed)](https://huggingface.co/Tonic/GaiaMiniMed) or duplicate this Space and use it locally or on 🤗HuggingFace. [Join me on Discord to build together](https://discord.gg/VqTxc76K3u)."
+
+examples = [  
+    ["Assistant is a public health and medical expert ready to help the user.", [{"user": "Hi there, I have a question!", "assistant": "My name is Gaia, I'm a health and sanitation expert ready to answer your medical questions."}],  
+    ["Assistant is a public health and medical expert ready to help the user.", [{"user": "What is the proper treatment for buccal herpes?", "assistant": None}]]  
+]
 
 additional_inputs=[
     gr.Textbox("", label="Optional system prompt"),
@@ -114,16 +123,19 @@ additional_inputs=[
     )
 ]
 
-with gr.Blocks() as demo:
-    title = "👋🏻Welcome to Tonic's GaiaMiniMed🦅⚕️Falcon Chat🚀"
-    description = "You can use this Space to test out the current model [(Tonic/GaiaMiniMed)](https://huggingface.co/Tonic/GaiaMiniMed) with chat memory optimized for falcon models or duplicate this Space and use it locally or on 🤗HuggingFace. [Join me on Discord to build together](https://discord.gg/VqTxc76K3u)."
-
-client = gr.Interface(
-    generate, 
-    examples=EXAMPLES,
-    additional_inputs=additional_inputs,
+iface = gr.Interface(
+    fn=falcon_bot.predict,
+    title=title,
+    description=description,
+    examples=examples,
+    inputs=[
+        gr.inputs.Textbox(label="System Prompt", type="text", lines=2),
+        gr.inputs.Textbox(label="User Message", type="text", lines=3),
+        gr.inputs.Textbox(label="Assistant Message", type="text", lines=2),
+    ] + additional_inputs,
+    outputs="text",
     theme="ParityError/Anime"
 )
 
-# Launch the Gradio interface
-client.launch(show_api=True)
+# Launch the Gradio interface for the Falcon model
+iface.launch()