Jofthomas's picture
Jofthomas HF staff
Update app.py
b799309 verified
import re
import streamlit as st
import requests
import pandas as pd
from io import StringIO
import plotly.graph_objs as go
from huggingface_hub import HfApi
from huggingface_hub.utils import RepositoryNotFoundError, RevisionNotFoundError
#from yall import create_yall
def place_holder_dataframe():
list_dict = [
{"gist_id":"mistralai/Mistral-7B-Instruct-v0.3",
"filename":"https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3/blob/main/README.md",
"url":"https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3",
"model_name":"Mistral-7B-Instruct-v0.3",
"model_id":"mistralai/Mistral-7B-Instruct-v0.3",
"Model":"Mistral-7B-Instruct-v0.3",
"Elo":1200,
"Undetected rate":0.27
},
{
"gist_id":"mistralai/Mixtral-8x22B-Instruct-v0.1",
"filename":"https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1/blob/main/README.md",
"url":"https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1",
"model_name":"Mixtral-8x22B-Instruct-v0.1",
"model_id":"mistralai/Mixtral-8x22B-Instruct-v0.1",
"Model":"Mixtral-8x22B-Instruct-v0.1",
"Elo":1950,
"Undetected rate":0.63
},
{
"gist_id":"mistralai/Mixtral-8x7B-Instruct-v0.1",
"filename":"https://huggingface.co/mistralai/Mixtral-8x7B-v0.1/blob/main/README.md",
"url":"https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1",
"model_name":"Mixtral-8x7B-Instruct-v0.1",
"model_id":"mistralai/Mixtral-8x7B-Instruct-v0.1",
"Model":"Mixtral-8x7B-Instruct-v0.1",
"Elo":1467,
"Undetected rate":0.41
}
]
df = pd.DataFrame(list_dict)
return df
def convert_markdown_table_to_dataframe(md_content):
"""
Converts markdown table to Pandas DataFrame, handling special characters and links,
extracts Hugging Face URLs, and adds them to a new column.
"""
# Remove leading and trailing | characters
cleaned_content = re.sub(r'\|\s*$', '', re.sub(r'^\|\s*', '', md_content, flags=re.MULTILINE), flags=re.MULTILINE)
# Create DataFrame from cleaned content
df = pd.read_csv(StringIO(cleaned_content), sep="\|", engine='python')
# Remove the first row after the header
df = df.drop(0, axis=0)
# Strip whitespace from column names
df.columns = df.columns.str.strip()
# Extract Hugging Face URLs and add them to a new column
model_link_pattern = r'\[(.*?)\]\((.*?)\)\s*\[.*?\]\(.*?\)'
df['URL'] = df['Model'].apply(lambda x: re.search(model_link_pattern, x).group(2) if re.search(model_link_pattern, x) else None)
# Clean Model column to have only the model link text
df['Model'] = df['Model'].apply(lambda x: re.sub(model_link_pattern, r'\1', x))
return df
@st.cache_data
def get_model_info(df):
api = HfApi()
# Initialize new columns for likes and tags
df['Likes'] = None
df['Tags'] = None
# Iterate through DataFrame rows
for index, row in df.iterrows():
model = row['Model'].strip()
try:
model_info = api.model_info(repo_id=str(model))
df.loc[index, 'Likes'] = model_info.likes
df.loc[index, 'Tags'] = ', '.join(model_info.tags)
except (RepositoryNotFoundError, RevisionNotFoundError):
df.loc[index, 'Likes'] = -1
df.loc[index, 'Tags'] = ''
return df
def create_bar_chart(df, category):
"""Create and display a bar chart for a given category."""
st.write(f"### {category} Scores")
# Sort the DataFrame based on the category score
sorted_df = df[['Model', category]].sort_values(by=category, ascending=True)
# Create the bar chart with a color gradient (using 'Viridis' color scale as an example)
fig = go.Figure(go.Bar(
x=sorted_df[category],
y=sorted_df['Model'],
orientation='h',
marker=dict(color=sorted_df[category], colorscale='Inferno')
))
# Update layout for better readability
fig.update_layout(
margin=dict(l=20, r=20, t=20, b=20)
)
# Adjust the height of the chart based on the number of rows in the DataFrame
st.plotly_chart(fig, use_container_width=True, height=35)
# Example usage:
# create_bar_chart(your_dataframe, 'Your_Category')
def main():
st.set_page_config(page_title="LLM Roleplay Leaderboard", layout="wide")
st.title("πŸ†πŸŽ­ LLM Roleplay Leaderboard")
st.markdown("LLM Roleplay Leaderboard that uses scores from the matou garou roleplay game πŸ πŸˆβ€.")
#content = create_yall()
tab1, tab2 = st.tabs(["πŸ†πŸŽ­ Leaderboard", "πŸ“ About"])
df = place_holder_dataframe()
with tab1:
if len(df)>0:
try:
df = df.sort_values(by='Elo', ascending=False)
# Add a search bar
search_query = st.text_input("Search models", "")
# Display the filtered DataFrame or the entire leaderboard
st.dataframe(
df[['Model', 'Elo', 'url', 'Undetected rate']],
use_container_width=True,
column_config={
"url": st.column_config.LinkColumn("url"),
},
hide_index=True,
)
# Filter the DataFrame based on the search query
if search_query:
df = df[df['Model'].str.contains(search_query, case=False)]
# Comparison between models
selected_models = st.multiselect('Select models to compare', df['Model'].unique())
comparison_df = df[df['Model'].isin(selected_models)]
st.dataframe(
comparison_df,
use_container_width=True,
column_config={
"url": st.column_config.LinkColumn("url"),
},
hide_index=True,
)
# Add a button to export data to CSV
if st.button("Export to CSV"):
# Export the DataFrame to CSV
csv_data = df.to_csv(index=False)
# Create a link to download the CSV file
st.download_button(
label="Download CSV",
data=csv_data,
file_name="leaderboard.csv",
key="download-csv",
help="Click to download the CSV file",
)
# Full-width plot for the first category
create_bar_chart(df, "Elo")
# Next two plots in two columns
col1, col2 = st.columns(2)
with col1:
create_bar_chart(df, "Undetected rate")
except Exception as e:
st.error("An error occurred while processing the markdown table.")
st.error(str(e))
else:
st.error("Failed to download the content from the URL provided.")
# About tab
with tab2:
st.markdown('''
### Roleplay Leaderboard
This space is here to present the results from the Matou-Garou space, where human and AI play a game of werewolf.
It is meant as a social experience to see if you would be able to detect if talking to an AI.
We also hope that this leaderboard can be used by video game creator in the future to select what model to select for LLM based NPCs
Popularized by [Teknium](https://huggingface.co/teknium) and [NousResearch](https://huggingface.co/NousResearch), this benchmark suite aggregates four benchmarks
Leaderboard copied from [Maxime Labonne](https://huggingface.co/mlabonne)
''')
if __name__ == "__main__":
main()