from dataclasses import dataclass from enum import Enum import json @dataclass class Task: benchmark: str metric: str col_name: str # Init: to update with your specific keys def create_task_list(): # task_key in the json file, metric_key in the json file, name to display in the leaderboard with open("src/datasets.json") as f: data = json.load(f) groups = [] names = [] for d in data: groups.append(d['group']) names.append(d['name']) groups = list(set(groups)) tasks = [] grouped_tasks = [] for name in names: tasks.append(Task(name, "metric_name", name)) for group in groups: grouped_tasks.append(Task(group, "metric_name", group)) return tasks, grouped_tasks # Your leaderboard name TITLE = """

Azerbaijani LLM Leaderboard

""" # What does your leaderboard evaluate? INTRODUCTION_TEXT = """ Welcome to Kapital Bank's Azerbaijani LLM Leaderboard. We use benchmarks in finance, banking, and general knowledge for accurate evaluations. 🚀 Submit Your Model 🚀 If you have a fine-tuned Azerbaijani LLM, submit it for evaluation! """ LLM_BENCHMARKS_TEXT = f""" ## How it works ## Reproducibility HERE we have about part """ LLM_DATASET_TEXT = f""" ## How it works ## Reproducibility HERE we have about part """ EVALUATION_QUEUE_TEXT = """ ## Some good practices before submitting a model ### 1) Make sure your model exists on hub. ### 2) Make sure your model is public. ## In case of model failure If your model is displayed in the `FAILED` category, its execution stopped. Make sure you have followed the above steps first. Please contact us if you are facing any trouble! """