File size: 5,302 Bytes
101c142 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 |
import streamlit as st
import pandas as pd
import numpy as np
from streamlit_echarts import st_echarts
from streamlit.components.v1 import html
# from PIL import Image
from app.show_examples import *
import pandas as pd
from typing import List
from model_information import get_dataframe
info_df = get_dataframe()
metrics_info = {
'wer': 'Word Error Rate (WER), a common metric for ASR evaluation. (The lower, the better)',
'llama3_70b_judge_binary': 'Binary evaluation using the LLAMA3-70B model, for tasks requiring a binary outcome. (0-100 based on score 0-1)',
'llama3_70b_judge': 'General evaluation using the LLAMA3-70B model, typically scoring based on subjective judgments. (0-100 based on score 0-5)',
'meteor': 'METEOR, a metric used for evaluating text generation, often used in translation or summarization tasks. (Sensitive to output length)',
'bleu': 'BLEU (Bilingual Evaluation Understudy), another text generation evaluation metric commonly used in machine translation. (Sensitive to output length)',
}
def sum_table_mulit_metrix(task_name, metrics_lists: List[str]):
for metrics in metrics_lists:
folder = f"./results/{metrics}/"
data_path = f'{folder}/{task_name.lower()}.csv'
chart_data = pd.read_csv(data_path).round(3)
selected_columns = [i for i in chart_data.columns if i != 'Model']
chart_data['Average'] = chart_data[selected_columns].mean(axis=1)
# new_dataset_name = dataset_name.replace('-', '_').lower()
st.markdown("""
<style>
.stMultiSelect [data-baseweb=select] span {
max-width: 800px;
font-size: 0.9rem;
background-color: #3C6478 !important; /* Background color for selected items */
color: white; /* Change text color */
back
}
</style>
""", unsafe_allow_html=True)
# remap model names
display_model_names = {key.strip() :val.strip() for key, val in zip(info_df['Original Name'], info_df['Proper Display Name'])}
chart_data['model_show'] = chart_data['Model'].map(lambda x: display_model_names.get(x, x))
models = st.multiselect("Please choose the model",
sorted(chart_data['model_show'].tolist()),
default = sorted(chart_data['model_show'].tolist()),
key=f"multiselect_{task_name}_{metrics}"
)
chart_data = chart_data[chart_data['model_show'].isin(models)].dropna(axis=0)
# chart_data = chart_data.sort_values(by=['Average'], ascending=True).dropna(axis=0)
if len(chart_data) == 0: return
# = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
'''
Show Table
'''
with st.container():
st.markdown(f'#### Overal Evaluation Results')
st.markdown(f'###### Evaluation Method: {metrics_info[metrics]}')
model_link = {key.strip(): val for key, val in zip(info_df['Proper Display Name'], info_df['Link'])}
chart_data['model_link'] = chart_data['model_show'].map(model_link)
tabel_columns = [i for i in chart_data.columns if i not in ['Model', 'model_show']]
column_to_front = 'Average'
new_order = [column_to_front] + [col for col in tabel_columns if col != column_to_front]
chart_data_table = chart_data[['model_show'] + new_order]
# Format numeric columns to 2 decimal places
chart_data_table[chart_data_table.columns[1]] = chart_data_table[chart_data_table.columns[1]].apply(lambda x: round(float(x), 3) if isinstance(float(x), (int, float)) else float(x))
if metrics in ['wer']:
ascend = True
else:
ascend= False
chart_data_table = chart_data_table.sort_values(
by=['Average'],
ascending=ascend
).reset_index(drop=True)
def highlight_first_element(x):
# Create a DataFrame with the same shape as the input
df_style = pd.DataFrame('', index=x.index, columns=x.columns)
# Apply background color to the first element in row 0 (df[0][0])
df_style.iloc[0, 1] = 'background-color: #b0c1d7; color: white'
return df_style
styled_df = chart_data_table.style.apply(
highlight_first_element, axis=None
)
st.dataframe(
styled_df,
column_config={
'model_show': 'Model',
chart_data_table.columns[1]: {'alignment': 'left'},
"model_link": st.column_config.LinkColumn(
"Model Link",
),
},
hide_index=True,
use_container_width=True
) |