Update app.py

app.py

@@ -1,38 +1,42 @@
 import streamlit as st
 import pandas as pd
-from huggingface_hub import HfApi
+from huggingface_hub import HfApi, ModelCard
 from huggingface_hub.utils import RepositoryNotFoundError, RevisionNotFoundError
 import re
 from io import StringIO
 from yall import create_yall
 import plotly.graph_objs as go
-from huggingface_hub import ModelCard
 
 def calculate_pages(df, items_per_page):
+    """Calculate the number of pages needed for pagination."""
     return -(-len(df) // items_per_page)  # Equivalent to math.ceil(len(df) / items_per_page)
 
 @st.cache_data
 def cached_model_info(_api, model):
+    """Fetch model information from the Hugging Face API and cache the result."""
     try:
         return _api.model_info(repo_id=str(model))
-    except (RepositoryNotFoundError, RevisionNotFoundError):
+    except (RepositoryNotFoundError, RevisionNotFoundError) as e:
+        st.error(f"Error fetching model info for {model}: {str(e)}")
         return None
 
 @st.cache_data
 def get_model_info(df):
+    """Get model information and update the DataFrame with likes and tags."""
     api = HfApi()
-
-    for index, row in df.iterrows():
-        model_info = cached_model_info(api, row['Model'].strip())
-        if model_info:
-            df.loc[index, 'Likes'] = model_info.likes
-            df.loc[index, 'Tags'] = ', '.join(model_info.tags)
-        else:
-            df.loc[index, 'Likes'] = -1
-            df.loc[index, 'Tags'] = ''
+    with st.spinner("Fetching model information..."):
+        for index, row in df.iterrows():
+            model_info = cached_model_info(api, row['Model'].strip())
+            if model_info:
+                df.loc[index, 'Likes'] = model_info.likes
+                df.loc[index, 'Tags'] = ', '.join(model_info.tags)
+            else:
+                df.loc[index, 'Likes'] = -1
+                df.loc[index, 'Tags'] = ''
     return df
 
 def convert_markdown_table_to_dataframe(md_content):
+    """Convert a markdown table to a pandas DataFrame."""
     cleaned_content = re.sub(r'\|\s*$', '', re.sub(r'^\|\s*', '', md_content, flags=re.MULTILINE), flags=re.MULTILINE)
     df = pd.read_csv(StringIO(cleaned_content), sep="\|", engine='python')
     df = df.drop(0, axis=0)
@@ -43,46 +47,57 @@ def convert_markdown_table_to_dataframe(md_content):
     return df
 
 def create_bar_chart(df, category):
+    """Create a horizontal bar chart for the specified category."""
     st.write(f"### {category} Scores")
     sorted_df = df[['Model', category]].sort_values(by=category, ascending=True)
     fig = go.Figure(go.Bar(
         x=sorted_df[category],
         y=sorted_df['Model'],
         orientation='h',
-        marker=dict(color=sorted_df[category], colorscale='Agsunset')
+        marker=dict(color=sorted_df[category], colorscale='Viridis'),
+        hoverinfo='x+y'
     ))
     fig.update_layout(
-        margin=dict(l=20, r=20, t=20, b=20)
+        margin=dict(l=20, r=20, t=20, b=20),
+        title=f"Leaderboard for {category} Scores"
     )
     st.plotly_chart(fig, use_container_width=True, height=len(df) * 35)
 
 def fetch_merge_configs(df):
+    """Fetch and save merge configurations for the top models."""
     df_sorted = df.sort_values(by='Average', ascending=False)
-    with open('/tmp/configurations.txt', 'a') as file:
-        for index, row in df_sorted.head(20).iterrows():
-            model_name = row['Model'].rstrip()
-            card = ModelCard.load(model_name)
-            file.write(f'Model Name: {model_name}\n')
-            file.write(f'Scores: {row["Average"]}\n')
-            file.write(f'AGIEval: {row["AGIEval"]}\n')
-            file.write(f'GPT4All: {row["GPT4All"]}\n')
-            file.write(f'TruthfulQA: {row["TruthfulQA"]}\n')
-            file.write(f'Bigbench: {row["Bigbench"]}\n')
-            file.write(f'Model Card: {card}\n')
-    with open('/tmp/configurations.txt', 'r') as file:
-        content = file.read()
-        matches = re.findall(r'yaml(.*?)```', content, re.DOTALL)
-    with open('/tmp/configurations2.txt', 'w') as file:
-        for row, match in zip(df_sorted[['Model', 'Average', 'AGIEval', 'GPT4All', 'TruthfulQA', 'Bigbench']].head(20).values, matches):
-            file.write(f'Model Name: {row[0]}\n')
-            file.write(f'Scores: {row[1]}\n')
-            file.write(f'AGIEval: {row[2]}\n')
-            file.write(f'GPT4All: {row[3]}\n')
-            file.write(f'TruthfulQA: {row[4]}\n')
-            file.write(f'Bigbench: {row[5]}\n')
-            file.write('yaml' + match + '```\n')
+    try:
+        with open('/tmp/configurations.txt', 'a') as file:
+            for index, row in df_sorted.head(20).iterrows():
+                model_name = row['Model'].rstrip()
+                try:
+                    card = ModelCard.load(model_name)
+                    file.write(f'Model Name: {model_name}\n')
+                    file.write(f'Scores: {row["Average"]}\n')
+                    file.write(f'AGIEval: {row["AGIEval"]}\n')
+                    file.write(f'GPT4All: {row["GPT4All"]}\n')
+                    file.write(f'TruthfulQA: {row["TruthfulQA"]}\n')
+                    file.write(f'Bigbench: {row["Bigbench"]}\n')
+                    file.write(f'Model Card: {card}\n')
+                except Exception as e:
+                    st.error(f"Error loading model card for {model_name}: {str(e)}")
+        with open('/tmp/configurations.txt', 'r') as file:
+            content = file.read()
+            matches = re.findall(r'yaml(.*?)```', content, re.DOTALL)
+        with open('/tmp/configurations2.txt', 'w') as file:
+            for row, match in zip(df_sorted[['Model', 'Average', 'AGIEval', 'GPT4All', 'TruthfulQA', 'Bigbench']].head(20).values, matches):
+                file.write(f'Model Name: {row[0]}\n')
+                file.write(f'Scores: {row[1]}\n')
+                file.write(f'AGIEval: {row[2]}\n')
+                file.write(f'GPT4All: {row[3]}\n')
+                file.write(f'TruthfulQA: {row[4]}\n')
+                file.write(f'Bigbench: {row[5]}\n')
+                file.write('yaml' + match + '```\n')
+    except Exception as e:
+        st.error(f"Error while fetching merge configs: {str(e)}")
 
 def main():
+    """Main function to set up the Streamlit app and display the leaderboard."""
     st.set_page_config(page_title="YALL - Yet Another LLM Leaderboard", layout="wide")
     st.title("🏆 YALL - Yet Another LLM Leaderboard")
     st.markdown("Leaderboard made with 🧐 [LLM AutoEval](https://github.com/mlabonne/llm-autoeval) using [Nous](https://huggingface.co/NousResearch) benchmark suite.")