frontend / src /display /about.py
MirakramAghalarov's picture
Testing Second table
dfc075f
raw
history blame
2.21 kB
from dataclasses import dataclass
from enum import Enum
@dataclass
class Task:
benchmark: str
metric: str
col_name: str
# Init: to update with your specific keys
class Tasks(Enum):
# task_key in the json file, metric_key in the json file, name to display in the leaderboard
task0 = Task("MMLU", "metric_name", "MMLU")
task1 = Task("Synthetic_QA", "metric_name", "Synthetic_QA")
task2 = Task("Support_MC", "metric_name", "Support_MC")
task3 = Task("Context_QA", "metric_name", "Context_QA")
task4 = Task("Banking_MC", "metric_name", "Banking_MC")
task5 = Task("ARC", "metric_name", "ARC")
task6 = Task("Binary_QA", "metric_name", "Binary_QA")
task7 = Task("ANL_Quad", "metric_name", "ANL_Quad")
class Tasks_Grouped(Enum):
task0 = Task("MMLU", "metric_name", "MMLU")
task1 = Task("Synthetic_QA", "metric_name", "Synthetic_QA")
task2 = Task("Support_MC", "metric_name", "Support_MC")
task3 = Task("Context_QA", "metric_name", "Context_QA")
task4 = Task("Banking_MC", "metric_name", "Banking_MC")
task5 = Task("ARC", "metric_name", "ARC")
task6 = Task("Binary_QA", "metric_name", "Binary_QA")
task7 = Task("ANL_Quad", "metric_name", "ANL_Quad")
# Your leaderboard name
TITLE = """<h1 align="center" id="space-title"> Azerbaijani LLM Leaderboard</h1>"""
# 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.
πŸš€ Submit Your Model πŸš€
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!
"""