core/utils/generate_problem.py

from llm.llm import LLM
from input.problem import problem_input
# from input.test_middle_result import problem_str, problem_analysis, selected_models, modeling_solution, modeling_solution, task_descriptions
from agent.problem_analysis import ProblemAnalysis
from agent.problem_modeling import ProblemModeling
from agent.task_decompse import TaskDecompose
from agent.task import Task
from agent.create_charts import Chart
from utils.utils import read_json_file, write_json_file, write_text_file, json_to_markdown
# from utils.convert_format import markdown_to_latex
import os
from prompt.template import PROBLEM_EXTRACT_PROMPT


config = {
    'problem_analysis_round': 1,
    'problem_modeling_round': 1,
    'task_formulas_round': 1,
    'tasknum': 4,
    'chart_num': 3,
    'model_name': 'chatgpt-4o-latest'
}

def run_batch(problem_path, config):
    # Initialize LLM
    llm = LLM(config['model_name'])
    
    # Get problem input
    problem_str, problem = problem_input(problem_path, llm)
    problem_name = os.path.splitext(os.path.basename(problem_path))[0]
    problem_type = os.path.splitext(os.path.basename(problem_path))[0].split('_')[-1]
    
    return {problem_name: problem}


if __name__ == "__main__":
    import glob
    # files = glob.glob('/Users/ann/Downloads/methmatical_paper_extraction/parse/2025_*/content/*.md')
    files = glob.glob('../data/actor_data/input/problem/2025_*')
    problems = read_json_file('../data/actor_data/output/problem_24.json')
    for file in files:
        problems.update(run_batch(file, config))

    write_json_file('../data/actor_data/output/problem_25.json', problems)
    print(problems.keys())