Spaces:
Running
Running
File size: 2,830 Bytes
65c6479 441cdc8 e608ddc 97d7225 65c6479 9c1b957 97d7225 541cf85 9c1b957 e608ddc 85b9042 e608ddc 65c6479 97d7225 85b9042 97d7225 541cf85 1e647ba 441cdc8 9c1b957 d0bceed 9c1b957 541cf85 9c1b957 1e647ba 9c1b957 74fdd72 9c1b957 cfb8d80 1e647ba cfb8d80 653c0f4 cfb8d80 1e647ba cfb8d80 4d3390f 441cdc8 9c1b957 97d7225 cfb8d80 97d7225 653c0f4 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 |
import gradio as gr
import pandas as pd
import os
from huggingface_hub import snapshot_download
from apscheduler.schedulers.background import BackgroundScheduler
from src.display.about import (
CITATION_BUTTON_LABEL,
CITATION_BUTTON_TEXT,
EVALUATION_QUEUE_TEXT,
INTRODUCTION_TEXT,
LLM_BENCHMARKS_TEXT,
TITLE,
)
from src.display.css_html_js import custom_css
from src.envs import API
from src.leaderboard.load_results import load_data
# clone / pull the lmeh eval data
TOKEN = os.environ.get("TOKEN", None)
RESULTS_REPO = f"SeaLLMs/SeaExam-results"
CACHE_PATH=os.getenv("HF_HOME", ".")
EVAL_RESULTS_PATH = os.path.join(CACHE_PATH, "eval-results")
print(EVAL_RESULTS_PATH)
snapshot_download(
repo_id=RESULTS_REPO, local_dir=EVAL_RESULTS_PATH, repo_type="dataset",
token=TOKEN
)
def restart_space():
API.restart_space(repo_id="SeaLLMs/SeaExam_leaderboard", token=TOKEN)
# Load the data from the csv file
csv_path = f'{EVAL_RESULTS_PATH}/SeaExam_results.csv'
df_m3exam, df_mmlu, df_avg = load_data(csv_path)
demo = gr.Blocks(css=custom_css)
with demo:
gr.HTML(TITLE)
# gr.Markdown(INTRODUCTION_TEXT, elem_classes="markdown-text")
with gr.Tabs(elem_classes="tab-buttons") as tabs:
with gr.TabItem("🏅 Overall", elem_id="llm-benchmark-Sum", id=0):
leaderboard_table = gr.components.Dataframe(
value=df_avg,
# value=leaderboard_df[
# [c.name for c in fields(AutoEvalColumn) if c.never_hidden]
# + shown_columns.value
# + [AutoEvalColumn.dummy.name]
# ],
# headers=[c.name for c in fields(AutoEvalColumn) if c.never_hidden] + shown_columns.value,
# datatype=TYPES,
# elem_id="leaderboard-table",
interactive=False,
visible=True,
# column_widths=["20%", "6%", "8%", "6%", "8%", "8%", "6%", "6%", "6%", "6%", "6%"],
)
with gr.TabItem("🏅 M3Exam", elem_id="llm-benchmark-M3Exam", id=1):
leaderboard_table = gr.components.Dataframe(
value=df_m3exam,
interactive=False,
visible=True,
)
with gr.TabItem("🏅 MMLU", elem_id="llm-benchmark-MMLU", id=2):
leaderboard_table = gr.components.Dataframe(
value=df_mmlu,
interactive=False,
visible=True,
)
with gr.TabItem("📝 About", elem_id="llm-benchmark-tab-table", id=3):
gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")
demo.launch()
scheduler = BackgroundScheduler()
scheduler.add_job(restart_space, "interval", seconds=1800)
scheduler.start()
demo.queue(default_concurrency_limit=40).launch(share=True)
|