Spaces:
Paused
Paused
File size: 1,836 Bytes
e60235b 969e123 41cd888 55af4cc 969e123 e60235b 969e123 e60235b 969e123 e60235b 97fcb64 969e123 97fcb64 |
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 |
import streamlit as st
import pandas as pd
import opendashboards.utils.utils as utils
def clean_data(df):
task_map = {
'question-answering': 'QA',
'summarization': 'Summarization',
'date-based question answering': 'Date QA',
'math': 'Math'
}
df['task'] = df.task.map(task_map)
return df.dropna(subset=df.filter(regex='completions|rewards').columns, how='any')
@st.cache_data
def explode_data(df):
list_cols = utils.get_list_col_lengths(df)
try:
return utils.explode_data(df, list(list_cols.keys())).apply(pd.to_numeric, errors='ignore')
except Exception as e:
st.error(f'Error exploding data with columns')
st.write(list_cols)
st.exception(e)
st.dataframe(df)
st.stop()
@st.cache_data
def completions(df_long, col):
return df_long[col].value_counts()
def run_event_data(df_runs, df, selected_runs):
st.markdown('#')
show_col1, show_col2 = st.columns(2)
show_runs = show_col1.checkbox('Show runs', value=True)
show_events = show_col2.checkbox('Show events', value=False)
if show_runs:
st.markdown(f'Wandb info for **{len(selected_runs)} selected runs**:')
st.dataframe(df_runs.loc[df_runs.id.isin(selected_runs)],
column_config={
"url": st.column_config.LinkColumn("URL"),
}
)
if show_events:
st.markdown(f'Raw events for **{len(selected_runs)} selected runs**:')
st.dataframe(df.head(50),
column_config={
"url": st.column_config.LinkColumn("URL"),
}
)
def highlight_row(row, expr, color='lightgrey', bg_color='white'):
return [f'background-color:{color}' if expr else f'background-color:{bg_color}'] * len(row) |