api_runner.py

import json
import os
from concurrent.futures import ThreadPoolExecutor, as_completed
from typing import Optional
from eval.eval import compare_query_results
import pandas as pd
from utils.pruning import prune_metadata_str
from utils.questions import prepare_questions_df
from utils.creds import db_creds_all, bq_project
from tqdm import tqdm
from time import time
import requests


def generate_prompt(
    prompt_file, question, db_name, instructions="", k_shot_prompt="", public_data=True
):
    with open(prompt_file, "r") as f:
        prompt = f.read()
    question_instructions = question + " " + instructions

    pruned_metadata_str = prune_metadata_str(
        question_instructions, db_name, public_data
    )
    prompt = prompt.format(
        user_question=question,
        instructions=instructions,
        table_metadata_string=pruned_metadata_str,
        k_shot_prompt=k_shot_prompt,
    )
    return prompt


def process_row(row, api_url, num_beams):
    start_time = time()
    r = requests.post(
        api_url,
        json={
            "prompt": row["prompt"],
            "n": 1,
            "use_beam_search": True,
            "best_of": num_beams,
            "temperature": 0,
            "stop": [";", "```"],
            "max_tokens": 600,
        },
    )
    end_time = time()
    generated_query = (
        r.json()["text"][0].split("```")[-1].split("```")[0].split(";")[0].strip() + ";"
    )

    row["generated_query"] = generated_query
    row["latency_seconds"] = end_time - start_time
    golden_query = row["query"]
    db_name = row["db_name"]
    db_type = row["db_type"]
    question = row["question"]
    query_category = row["query_category"]
    exact_match = correct = 0

    try:
        exact_match, correct = compare_query_results(
            query_gold=golden_query,
            query_gen=generated_query,
            db_name=db_name,
            db_type=db_type,
            db_creds=db_creds_all[row["db_type"]],
            question=question,
            query_category=query_category,
        )
        row["exact_match"] = int(exact_match)
        row["correct"] = int(correct)
        row["error_msg"] = ""
    except Exception as e:
        row["error_db_exec"] = 1
        row["error_msg"] = f"QUERY EXECUTION ERROR: {e}"

    return row


def run_api_eval(args):
    # get params from args
    questions_file = args.questions_file
    prompt_file_list = args.prompt_file
    num_questions = args.num_questions
    public_data = not args.use_private_data
    api_url = args.url
    output_file_list = args.output_file
    k_shot = args.k_shot
    num_beams = args.num_beams
    max_workers = args.parallel_threads
    db_type = args.db_type

    # get questions
    print("Preparing questions...")
    print(
        f"Using {'all' if num_questions is None else num_questions} question(s) from {questions_file}"
    )
    df = prepare_questions_df(questions_file, db_type, num_questions, k_shot)

    for prompt_file, output_file in zip(prompt_file_list, output_file_list):
        # create a prompt for each question
        df["prompt"] = df[
            ["question", "db_name", "instructions", "k_shot_prompt"]
        ].apply(
            lambda row: generate_prompt(
                prompt_file,
                row["question"],
                row["db_name"],
                row["instructions"],
                row["k_shot_prompt"],
                public_data,
            ),
            axis=1,
        )

        print(f"Questions prepared\nNow loading model...")
        # initialize tokenizer and model
        total_tried = 0
        total_correct = 0
        output_rows = []

        with ThreadPoolExecutor(max_workers=max_workers) as executor:
            futures = []
            for row in df.to_dict("records"):
                futures.append(executor.submit(process_row, row, api_url, num_beams))

            with tqdm(as_completed(futures), total=len(futures)) as pbar:
                for f in pbar:
                    row = f.result()
                    output_rows.append(row)
                    if row["correct"]:
                        total_correct += 1
                    total_tried += 1
                    pbar.update(1)
                    pbar.set_description(
                        f"Correct so far: {total_correct}/{total_tried} ({100*total_correct/total_tried:.2f}%)"
                    )

        output_df = pd.DataFrame(output_rows)
        del output_df["prompt"]
        print(output_df.groupby("query_category")[["exact_match", "correct"]].mean())
        output_df = output_df.sort_values(by=["db_name", "query_category", "question"])
        # get directory of output_file and create if not exist
        output_dir = os.path.dirname(output_file)
        if not os.path.exists(output_dir):
            os.makedirs(output_dir)
        try:
            output_df.to_csv(output_file, index=False, float_format="%.2f")
        except:
            output_df.to_pickle(output_file)

        # save to BQ
        if args.bq_table is not None:
            run_name = output_file.split("/")[-1].split(".")[0]
            output_df["run_name"] = run_name
            output_df["run_time"] = pd.Timestamp.now()
            output_df["run_params"] = json.dumps(vars(args))
            print(f"Saving to BQ table {args.bq_table} with run_name {run_name}")
            try:
                if bq_project is not None and bq_project != "":
                    output_df.to_gbq(
                        destination_table=args.bq_table,
                        project_id=bq_project,
                        if_exists="append",
                        progress_bar=False,
                    )
                else:
                    print("No BQ project id specified, skipping save to BQ")
            except Exception as e:
                print(f"Error saving to BQ: {e}")