adding new needed data source for pearl agents

app.py

@@ -27,7 +27,11 @@ from tabs.trader_plots import (
     plot_total_bet_amount,
     plot_active_traders,
 )
-from tabs.agent_graphs import plot_rolling_average_dune, plot_rolling_average_roi
+from tabs.agent_graphs import (
+    plot_rolling_average_dune,
+    plot_rolling_average_roi,
+    plot_weekly_average_roi,
+)
 
 from tabs.daily_graphs import (
     get_current_week_data,
@@ -151,7 +155,14 @@ def load_all_data():
         repo_type="dataset",
     )
     df9 = pd.read_parquet(daa_pearl_df)
-    return df1, df2, df3, df4, df5, df6, df7, df8, df9
+    # Read pearl_agents.parquet
+    pearl_agents_df = hf_hub_download(
+        repo_id="valory/Olas-predict-dataset",
+        filename="pearl_agents.parquet",
+        repo_type="dataset",
+    )
+    df10 = pd.read_parquet(pearl_agents_df)
+    return df1, df2, df3, df4, df5, df6, df7, df8, df9, df10
 
 
 def prepare_data():
@@ -166,6 +177,7 @@ def prepare_data():
         all_mech_calls,
         daa_qs_df,
         daa_pearl_df,
+        pearl_agents_df,
     ) = load_all_data()
     all_trades["creation_timestamp"] = all_trades["creation_timestamp"].dt.tz_convert(
         "UTC"
@@ -243,6 +255,7 @@ def prepare_data():
         all_mech_calls,
         daa_qs_df,
         daa_pearl_df,
+        pearl_agents_df,
     )
 
 
@@ -256,6 +269,7 @@ def prepare_data():
     all_mech_calls,
     daa_qs_df,
     daa_pearl_df,
+    pearl_agents_df,
 ) = prepare_data()
 
 retention_df = prepare_retention_dataset(
@@ -559,10 +573,19 @@ with demo:
             with gr.Row():
                 pearl_rolling_avg_plot = plot_rolling_average_roi(
                     weekly_roi_df=weekly_metrics_by_market_creator,
-                    market_creator="pearl",
+                    pearl_agents=pearl_agents_df,
+                )
+            with gr.Row():
+                gr.Markdown("# Average weekly ROI for Pearl agents")
+            with gr.Row():
+                gr.Markdown(
+                    "This graph shows the average weekly ROI for Pearl agents. The data is based on the latest DAA results."
                 )
             with gr.Row():
-                gr.Markdown("# Average weekly ROI for Pearl agents (WIP)")
+                weekly_avg_roi_plot = plot_weekly_average_roi(
+                    weekly_roi_df=weekly_metrics_by_market_creator,
+                    pearl_agents=pearl_agents_df,
+                )
 
         with gr.TabItem("🪝 Retention metrics (WIP)"):
             with gr.Row():

tabs/agent_graphs.py

@@ -7,7 +7,7 @@ import plotly.express as px
 def plot_rolling_average_dune(
     daa_df: pd.DataFrame,
 ) -> gr.Plot:
-    """Function to plot the rolling average of daily active traders"""
+    """Function to plot the rolling average of daily active agents"""
 
     fig = px.bar(
         daa_df,
@@ -28,7 +28,7 @@ def plot_rolling_average(
     daa_df: pd.DataFrame,
     market_creator: str = None,
 ) -> gr.Plot:
-    """Function to plot the rolling average of daily active traders by markets"""
+    """Function to plot the rolling average of daily active agents"""
     if market_creator is not None:
         filtered_traders_df = daa_df.loc[daa_df["market_creator"] == market_creator]
         rolling_avg_df = get_sevenday_rolling_average(filtered_traders_df)
@@ -77,15 +77,24 @@ def get_sevenday_rolling_average(daa_df: pd.DataFrame) -> pd.DataFrame:
 
 
 def plot_rolling_average_roi(
-    weekly_roi_df: pd.DataFrame, daa_pearl_df: pd.DataFrame
+    weekly_roi_df: pd.DataFrame, pearl_agents: pd.DataFrame
 ) -> gr.Plot:
     """Function to plot the rolling average of ROI for pearl agents"""
     # Get the list of unique addresses from the daa_pearl_df
-    unique_addresses = daa_pearl_df["trader_address"].unique()
+    unique_addresses = pearl_agents["safe_address"].unique()
     # Filter the weekly_roi_df to include only those addresses
     filtered_weekly_roi_df = weekly_roi_df[
         weekly_roi_df["trader_address"].isin(unique_addresses)
     ]
+    # Select only the columns: "roi", "month_year_week", "trader_address"
+    filtered_weekly_roi_df = filtered_weekly_roi_df[
+        ["roi", "month_year_week", "trader_address"]
+    ].copy()
+    # Remove duplicates
+    filtered_weekly_roi_df = filtered_weekly_roi_df.drop_duplicates(
+        subset=["month_year_week", "trader_address"]
+    )
+
     # Get the 2-week rolling average of ROI
     rolling_avg_roi_df = get_twoweeks_rolling_average_roi(filtered_weekly_roi_df)
     print(rolling_avg_roi_df.head())
@@ -125,3 +134,59 @@ def get_twoweeks_rolling_average_roi(weekly_roi_df: pd.DataFrame) -> pd.DataFram
     )
     trader_rolling_avg_roi.rename(columns={"roi": "rolling_avg_roi"}, inplace=True)
     return trader_rolling_avg_roi
+
+
+def get_weekly_average_roi(weekly_roi_df: pd.DataFrame) -> pd.DataFrame:
+    """Function to get the weekly average ROI for pearl agents"""
+    # Create a local copy of the dataframe
+    local_df = weekly_roi_df.copy()
+
+    # Convert string dates to datetime
+    local_df["month_year_week"] = pd.to_datetime(
+        local_df["month_year_week"], format="%b-%d-%Y"
+    )
+
+    # Group by month_year_week and market_creator, then calculate the mean ROI
+    weekly_avg_roi = (
+        local_df.groupby(["month_year_week"], sort=False)["roi"].mean().reset_index()
+    )
+    return weekly_avg_roi
+
+
+def plot_weekly_average_roi(
+    weekly_roi_df: pd.DataFrame, pearl_agents: pd.DataFrame
+) -> gr.Plot:
+    """Function to plot the weekly average of ROI for pearl agents"""
+    # Get the list of unique addresses from the daa_pearl_df
+    unique_addresses = pearl_agents["safe_address"].unique()
+    # Filter the weekly_roi_df to include only those addresses
+    filtered_weekly_roi_df = weekly_roi_df[
+        weekly_roi_df["trader_address"].isin(unique_addresses)
+    ]
+    # Select only the columns: "roi", "month_year_week", "trader_address"
+    filtered_weekly_roi_df = filtered_weekly_roi_df[
+        ["roi", "month_year_week", "trader_address"]
+    ].copy()
+    # Remove duplicates
+    filtered_weekly_roi_df = filtered_weekly_roi_df.drop_duplicates(
+        subset=["month_year_week", "trader_address"]
+    )
+    # Get the weekly average ROI
+    weekly_avg_roi_df = get_weekly_average_roi(filtered_weekly_roi_df)
+    # plot the weekly average ROI
+    print(weekly_avg_roi_df.head())
+    # Ensure 'month_year_week' is a column, not an index
+    if "month_year_week" not in weekly_avg_roi_df.columns:
+        weekly_avg_roi_df = weekly_avg_roi_df.reset_index()
+    fig = px.line(
+        weekly_avg_roi_df,
+        x="month_year_week",
+        y="roi",
+    )
+    fig.update_layout(
+        xaxis_title="Week",
+        yaxis_title="Weekly average ROI for pearl agents",
+    )
+    return gr.Plot(
+        value=fig,
+    )