Spaces:
Running
Running
File size: 2,851 Bytes
71b62ee aadfcea 98a40e2 22e8484 71b62ee aadfcea 89d0d4b 71b62ee 22e8484 129288f 22e8484 129288f 22e8484 129288f 71b62ee 8752ce9 22e8484 8752ce9 22e8484 71b62ee aadfcea 71b62ee |
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 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 |
from apscheduler.schedulers.background import BackgroundScheduler
import gradio as gr
import pandas as pd
# Load leaderboard data
leaderboard_df = pd.read_csv('leaderboard.csv')
leaderboard_auto_df = pd.read_csv('leaderboard_auto.csv')
# Constants
TITLE = "Financial Natural Language Understanding and Prediction Evaluation Benchmark (FLARE) Leaderboard"
INTRODUCTION_TEXT = "The leaderboard shows the performance of various models in financial natural language understanding and prediction tasks."
COLS = [
("Model", "str"),
("FPB-acc", "number"),
("FPB-F1", "number"),
("FiQA-SA-F1", "number"),
("Headline-AvgF1", "number"),
("NER-EntityF1", "number"),
("FinQA-EmAcc", "number"),
("ConvFinQA-EmAcc", "number"),
("BigData22-Acc", "number"),
("BigData22-MCC", "number"),
("ACL18-Acc", "number"),
("ACL18-MCC", "number"),
("CIKM18-Acc", "number"),
("CIKM18-MCC", "number")
]
COLS_AUTO = [
("Model", "str"),
("FPB-acc", "number"),
("FPB-F1", "number"),
("FPB-missing", "number"),
("FiQA-SA-F1", "number"),
("FiQA-SA-missing", "number"),
("Headline-AvgF1", "number"),
("NER-EntityF1", "number"),
("FinQA-EmAcc", "number"),
("BigData22-Acc", "number"),
("BigData22-MCC", "number"),
("BigData22-missing", "number"),
("ACL18-Acc", "number"),
("ACL18-MCC", "number"),
("ACL18-missing", "number"),
("CIKM18-Acc", "number"),
("CIKM18-MCC", "number"),
("CIKM18-missing", "number"),
("FOMC-acc", "number"),
("FOMC-F1", "number"),
("FOMC-missing", "number"),
("FinerOrd-EntityF1", "number"),
("FinerOrd-F1", "number"),
("German-Acc", "number"),
("German-MCC", "number"),
("German-missing", "number"),
("Australian-Acc", "number"),
("Australian-MCC", "number"),
("Australian-missing", "number")
]
TYPES = [col_type for _, col_type in COLS]
TYPES_AUTO = [col_type for _, col_type in COLS_AUTO]
def create_leaderboard_table(df, headers, types):
return gr.components.Dataframe(
value=df.values.tolist(),
headers=[col_name for col_name, _ in headers],
datatype=types,
max_rows=10,
)
def launch_gradio():
iface = gr.Interface(
fn=dummy_fn,
layout="tab",
inputs=None,
outputs=None,
title=TITLE,
description=INTRODUCTION_TEXT,
tabs=["Self-hosted evaluation", "Automatic evaluation"],
)
with iface.tab("Self-hosted evaluation"):
create_leaderboard_table(leaderboard_df, COLS, TYPES)
with iface.tab("Automatic evaluation"):
create_leaderboard_table(leaderboard_auto_df, COLS_AUTO, TYPES_AUTO)
iface.launch()
scheduler = BackgroundScheduler()
scheduler.add_job(launch_gradio, "interval", seconds=3600)
scheduler.start()
# Launch immediately
launch_gradio()
|