app.py

import os
from threading import Event, Thread
from transformers import (
    AutoModelForCausalLM,
    AutoTokenizer,
    StoppingCriteria,
    StoppingCriteriaList,
    TextIteratorStreamer,
)
import gradio as gr
import torch
import sqlparse

model_name = os.getenv("HF_MODEL_NAME", None)
tok = AutoTokenizer.from_pretrained(model_name)

max_new_tokens = 1024

print(f"Starting to load the model {model_name}")

m = AutoModelForCausalLM.from_pretrained(
    model_name,
    device_map=0,
    #load_in_8bit=True,
    cache_dir="/hf_cache",
)

m.config.pad_token_id = m.config.eos_token_id
m.generation_config.pad_token_id = m.config.eos_token_id

stop_tokens = [";", "###", "Result"]
stop_token_ids = tok.convert_tokens_to_ids(stop_tokens)

print(f"Successfully loaded the model {model_name} into memory")

class StopOnTokens(StoppingCriteria):
    def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTensor, **kwargs) -> bool:
        for stop_id in stop_token_ids:
            if input_ids[0][-1] == stop_id:
                return True
        return False

def bot(input_message: str, db_info="", temperature=0.1, top_p=0.9, top_k=0, repetition_penalty=1.08):
    stop = StopOnTokens()

    # Format the user's input message
    messages = f"Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n\n### Instruction:\n\nConvert text to sql: {input_message} {db_info}\n\n### Response:\n\n"

    input_ids = tok(messages, return_tensors="pt").input_ids
    input_ids = input_ids.to(m.device)
    streamer = TextIteratorStreamer(tok, timeout=100.0, skip_prompt=True, skip_special_tokens=True)
    generate_kwargs = dict(
        input_ids=input_ids,
        max_new_tokens=max_new_tokens,
        temperature=temperature,
        do_sample=temperature > 0.0,
        top_p=top_p,
        top_k=top_k,
        repetition_penalty=repetition_penalty,
        streamer=streamer,
        stopping_criteria=StoppingCriteriaList([stop]),
    )

    stream_complete = Event()

    def generate_and_signal_complete():
        m.generate(**generate_kwargs)
        stream_complete.set()

    t1 = Thread(target=generate_and_signal_complete)
    t1.start()

    partial_text = ""
    for new_text in streamer:
        partial_text += new_text

    # Split the text by "|", and get the last element in the list which should be the final query
    try:
        final_query = partial_text.split("|")[1].strip()
    except Exception:
        final_query = partial_text

    try:
        # Attempt to format SQL query using sqlparse
        formatted_query = sqlparse.format(final_query, reindent=True, keyword_case='upper')
    except Exception:
        # If formatting fails, use the original, unformatted query
        formatted_query = final_query

    # Convert SQL to markdown (not required, but just to show how to use the markdown module)
    final_query_markdown = f"{formatted_query}"
    return final_query_markdown

with gr.Blocks(theme='gradio/soft') as demo:
    header = gr.HTML("""
        <h1 style="text-align: center">SQL Skeleton WizardCoder Demo</h1>
        <h3 style="text-align: center">🧙‍♂️ Generate SQL queries from Natural Language 🧙‍♂️</h3>
    """)

    output_box = gr.Code(label="Generated SQL", lines=2, interactive=True)
    input_text = gr.Textbox(lines=3, placeholder='Write your question here...', label='NL Input')
    db_info = gr.Textbox(lines=4, placeholder='Example: | table_01 : column_01 , column_02 | table_02 : column_01 , column_02 | ...', label='Database Info')

    with gr.Accordion("Hyperparameters", open=False):
        temperature = gr.Slider(label="Temperature", minimum=0.0, maximum=1.0, value=0.5, step=0.1)
        top_p = gr.Slider(label="Top-p (nucleus sampling)", minimum=0.0, maximum=1.0, value=0.9, step=0.01)
        top_k = gr.Slider(label="Top-k", minimum=0, maximum=200, value=0, step=1)
        repetition_penalty = gr.Slider(label="Repetition Penalty", minimum=1.0, maximum=2.0, value=1.08, step=0.01)
        
    run_button = gr.Button("Generate SQL", variant="primary")
    
    with gr.Accordion("Examples", open=True):
        examples = gr.Examples([
            ["What is the average, minimum, and maximum age for all French singers?", "| stadium : stadium_id , location , name , capacity , highest , lowest , average | singer : singer_id , name , country , song_name , song_release_year , age , is_male | concert : concert_id , concert_name , theme , stadium_id , year | singer_in_concert : concert_id , singer_id | concert.stadium_id = stadium.stadium_id | singer_in_concert.singer_id = singer.singer_id | singer_in_concert.concert_id = concert.concert_id |"],
            ["Show location and name for all stadiums with a capacity between 5000 and 10000.", "| stadium : stadium_id , location , name , capacity , highest , lowest , average | singer : singer_id , name , country , song_name , song_release_year , age , is_male | concert : concert_id , concert_name , theme , stadium_id , year | singer_in_concert : concert_id , singer_id | concert.stadium_id = stadium.stadium_id | singer_in_concert.singer_id = singer.singer_id | singer_in_concert.concert_id = concert.concert_id |"],
            ["What are the number of concerts that occurred in the stadium with the largest capacity ?", "| stadium : stadium_id , location , name , capacity , highest , lowest , average | singer : singer_id , name , country , song_name , song_release_year , age , is_male | concert : concert_id , concert_name , theme , stadium_id , year | singer_in_concert : concert_id , singer_id | concert.stadium_id = stadium.stadium_id | singer_in_concert.singer_id = singer.singer_id | singer_in_concert.concert_id = concert.concert_id |"],
            ["How many male singers performed in concerts in the year 2023?", "| stadium : stadium_id , location , name , capacity , highest , lowest , average | singer : singer_id , name , country , song_name , song_release_year , age , is_male | concert : concert_id , concert_name , theme , stadium_id , year | singer_in_concert : concert_id , singer_id | concert.stadium_id = stadium.stadium_id | singer_in_concert.singer_id = singer.singer_id | singer_in_concert.concert_id = concert.concert_id |"],
            ["List the names of all singers who performed in a concert with the theme 'Rock'", "| stadium : stadium_id , location , name , capacity , highest , lowest , average | singer : singer_id , name , country , song_name , song_release_year , age , is_male | concert : concert_id , concert_name , theme , stadium_id , year | singer_in_concert : concert_id , singer_id | concert.stadium_id = stadium.stadium_id | singer_in_concert.singer_id = singer.singer_id | singer_in_concert.concert_id = concert.concert_id |"]
        ], inputs=[input_text, db_info, temperature, top_p, top_k, repetition_penalty], fn=bot)

    bitsandbytes_model = "richardr1126/spider-skeleton-wizard-coder-8bit"
    merged_model = "richardr1126/spider-skeleton-wizard-coder-merged"
    initial_model = "WizardLM/WizardCoder-15B-V1.0"
    finetuned_model = "richardr1126/spider-skeleton-wizard-coder-qlora"
    dataset = "richardr1126/spider-skeleton-context-instruct"
    
    footer = gr.HTML(f"""
        <p>🛠️ If you want you can <strong>duplicate this Space</strong>, then change the HF_MODEL_REPO spaces env varaible to use any Transformers model.</p>
        <p>🌐 Leveraging the <a href='https://huggingface.co/{bitsandbytes_model}'><strong>bitsandbytes 8-bit version</strong></a> of <a href='https://huggingface.co/{merged_model}'><strong>{merged_model}</strong></a> model.</p>
        <p>🔗 How it's made: <a href='https://huggingface.co/{initial_model}'><strong>{initial_model}</strong></a> was finetuned to create <a href='https://huggingface.co/{finetuned_model}'><strong>{finetuned_model}</strong></a>, then merged together to create <a href='https://huggingface.co/{merged_model}'><strong>{merged_model}</strong></a>.</p>
        <p>📉 Fine-tuning was performed using QLoRA techniques on the <a href='https://huggingface.co/datasets/{dataset}'><strong>{dataset}</strong></a> dataset. You can view training metrics on the <a href='https://huggingface.co/{finetuned_model}'><strong>QLoRa adapter HF Repo</strong></a>.</p>
    """)


    run_button.click(fn=bot, inputs=[input_text, db_info, temperature, top_p, top_k, repetition_penalty], outputs=output_box, api_name="txt2sql")

demo.queue(concurrency_count=1, max_size=10).launch()