import gradio as gr import pandas as pd import logging from scripts.utils import prepare_data from scripts.metrics import ( compute_weekly_metrics_by_market_creator, compute_daily_metrics_by_market_creator, compute_winning_metrics_by_trader, ) from scripts.retention_metrics import ( prepare_retention_dataset, calculate_wow_retention_by_type, calculate_cohort_retention, ) from tabs.trader_plots import ( plot_trader_metrics_by_agent_categories, default_trader_metric, trader_metric_choices, get_metrics_text, plot_winning_metric_per_trader, get_interpretation_text, plot_total_bet_amount, plot_active_traders, ) 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, plot_daily_metrics, trader_daily_metric_choices, default_daily_metric, ) from scripts.utils import get_traders_family from tabs.market_plots import ( plot_kl_div_per_market, plot_total_bet_amount_per_trader_per_market, ) from tabs.retention_plots import ( plot_wow_retention_by_type, plot_cohort_retention_heatmap, ) def get_logger(): logger = logging.getLogger(__name__) logger.setLevel(logging.DEBUG) # stream handler and formatter stream_handler = logging.StreamHandler() stream_handler.setLevel(logging.DEBUG) formatter = logging.Formatter( "%(asctime)s - %(name)s - %(levelname)s - %(message)s" ) stream_handler.setFormatter(formatter) logger.addHandler(stream_handler) return logger logger = get_logger() ( traders_data, closed_markets, daily_info, unknown_traders, raw_retention_df, active_traders, all_mech_calls, daa_qs_df, daa_pearl_df, weekly_avg_roi_pearl_agents, two_weeks_avg_roi_pearl_agents, traders_weekly_metrics_df, ) = prepare_data() retention_df = prepare_retention_dataset( retention_df=raw_retention_df, unknown_df=unknown_traders ) print("max date of retention df") print(max(retention_df.creation_timestamp)) demo = gr.Blocks() # TODO classify traders in the weekly metrics dataframe by agent type (Pearl, QS, and all) weekly_unknown_trader_metrics_by_market_creator = None if len(unknown_traders) > 0: weekly_unknown_trader_metrics_by_market_creator = ( compute_weekly_metrics_by_market_creator( traders_data=unknown_traders, all_mech_calls=None, trader_filter=None, unknown_trader=True, ) ) # just for all traders weekly_winning_metrics = compute_winning_metrics_by_trader( traders_data=traders_data, unknown_info=unknown_traders ) weekly_winning_metrics_olas = compute_winning_metrics_by_trader( traders_data=traders_data, unknown_info=unknown_traders, trader_filter="Olas" ) weekly_non_olas_winning_metrics = pd.DataFrame() if len(traders_data.loc[traders_data["staking"] == "non_Olas"]) > 0: weekly_non_olas_winning_metrics = compute_winning_metrics_by_trader( traders_data=traders_data, unknown_info=unknown_traders, trader_filter="non_Olas", ) with demo: gr.HTML("

Traders monitoring dashboard

") gr.Markdown("This app shows the weekly performance of the traders in Olas Predict.") with gr.Tabs(): with gr.TabItem("🔥 Weekly metrics (WIP)"): with gr.Row(): gr.Markdown("# Weekly metrics for all traders") with gr.Row(): trader_details_selector = gr.Dropdown( label="Select a weekly trader metric", choices=trader_metric_choices, value=default_trader_metric, ) with gr.Row(): with gr.Column(scale=3): trader_markets_plot = plot_trader_metrics_by_agent_categories( metric_name=default_trader_metric, traders_df=traders_weekly_metrics_df, ) with gr.Column(scale=1): trade_details_text = get_metrics_text(trader_type=None) def update_trader_details(trader_detail): return plot_trader_metrics_by_agent_categories( metric_name=trader_detail, traders_df=traders_weekly_metrics_df, ) trader_details_selector.change( update_trader_details, inputs=trader_details_selector, outputs=trader_markets_plot, ) # if len(weekly_non_olas_metrics_by_market_creator) > 0: # # Non-Olas traders graph # with gr.Row(): # gr.Markdown("# Weekly metrics of Non-Olas traders") # with gr.Row(): # trader_no_details_selector = gr.Dropdown( # label="Select a weekly trader metric", # choices=trader_metric_choices, # value=default_trader_metric, # ) # with gr.Row(): # with gr.Column(scale=3): # trader_no_markets_plot = plot_trader_metrics_by_market_creator( # metric_name=default_trader_metric, # traders_df=weekly_non_olas_metrics_by_market_creator, # ) # with gr.Column(scale=1): # trade_details_text = get_metrics_text(trader_type="non_Olas") # def update_no_trader_details(trader_detail): # return plot_trader_metrics_by_market_creator( # metric_name=trader_detail, # traders_df=weekly_non_olas_metrics_by_market_creator, # ) # trader_no_details_selector.change( # update_no_trader_details, # inputs=trader_no_details_selector, # outputs=trader_no_markets_plot, # ) # Unknown traders graph # if weekly_unknown_trader_metrics_by_market_creator is not None: # with gr.Row(): # gr.Markdown("# Weekly metrics of Unclassified traders") # with gr.Row(): # trader_u_details_selector = gr.Dropdown( # label="Select a weekly trader metric", # choices=trader_metric_choices, # value=default_trader_metric, # ) # with gr.Row(): # with gr.Column(scale=3): # trader_u_markets_plot = plot_trader_metrics_by_agent_categories( # metric_name=default_trader_metric, # traders_df=weekly_unknown_trader_metrics_by_market_creator, # ) # with gr.Column(scale=1): # trade_details_text = get_metrics_text( # trader_type="unclassified" # ) # def update_u_trader_details(trader_detail): # return plot_trader_metrics_by_agent_categories( # metric_name=trader_detail, # traders_df=weekly_unknown_trader_metrics_by_market_creator, # ) # trader_u_details_selector.change( # update_u_trader_details, # inputs=trader_u_details_selector, # outputs=trader_u_markets_plot, # ) with gr.TabItem("📅 Daily metrics (WIP)"): live_trades_current_week = get_current_week_data(trades_df=daily_info) if len(live_trades_current_week) > 0: live_metrics_by_market_creator = ( compute_daily_metrics_by_market_creator( live_trades_current_week, trader_filter=None, live_metrics=True ) ) else: live_metrics_by_market_creator = pd.DataFrame() with gr.Row(): gr.Markdown("# Daily live metrics for all trades") with gr.Row(): trade_live_details_selector = gr.Dropdown( label="Select a daily live metric", choices=trader_daily_metric_choices, value=default_daily_metric, ) with gr.Row(): with gr.Column(scale=3): trade_live_details_plot = plot_daily_metrics( metric_name=default_daily_metric, trades_df=live_metrics_by_market_creator, ) with gr.Column(scale=1): trade_details_text = get_metrics_text(daily=True) def update_trade_live_details(trade_detail, trade_live_details_plot): new_a_plot = plot_daily_metrics( metric_name=trade_detail, trades_df=live_metrics_by_market_creator ) return new_a_plot trade_live_details_selector.change( update_trade_live_details, inputs=[trade_live_details_selector, trade_live_details_plot], outputs=[trade_live_details_plot], ) # Olas traders with gr.Row(): gr.Markdown("# Daily live metrics for 🌊 Olas traders") with gr.Row(): o_trader_live_details_selector = gr.Dropdown( label="Select a daily live metric", choices=trader_daily_metric_choices, value=default_daily_metric, ) with gr.Row(): with gr.Column(scale=3): o_trader_live_details_plot = plot_daily_metrics( metric_name=default_daily_metric, trades_df=live_metrics_by_market_creator, trader_filter="Olas", ) with gr.Column(scale=1): trade_details_text = get_metrics_text(daily=True) def update_a_trader_live_details(trade_detail, a_trader_live_details_plot): o_trader_plot = plot_daily_metrics( metric_name=trade_detail, trades_df=live_metrics_by_market_creator, trader_filter="Olas", ) return o_trader_plot o_trader_live_details_selector.change( update_a_trader_live_details, inputs=[o_trader_live_details_selector, o_trader_live_details_plot], outputs=[o_trader_live_details_plot], ) with gr.Row(): gr.Markdown("# Daily live metrics for Non-Olas traders") with gr.Row(): no_trader_live_details_selector = gr.Dropdown( label="Select a daily live metric", choices=trader_daily_metric_choices, value=default_daily_metric, ) with gr.Row(): with gr.Column(scale=3): no_trader_live_details_plot = plot_daily_metrics( metric_name=default_daily_metric, trades_df=live_metrics_by_market_creator, trader_filter="non_Olas", ) with gr.Column(scale=1): trade_details_text = get_metrics_text(daily=True) def update_na_trader_live_details( trade_detail, no_trader_live_details_plot ): no_trader_plot = plot_daily_metrics( metric_name=trade_detail, trades_df=live_metrics_by_market_creator, trader_filter="non_Olas", ) return no_trader_plot no_trader_live_details_selector.change( update_na_trader_live_details, inputs=[no_trader_live_details_selector, no_trader_live_details_plot], outputs=[no_trader_live_details_plot], ) with gr.TabItem(" Agent metrics"): with gr.Row(): gr.Markdown(" # Daily active Pearl agents") with gr.Row(): rolling_avg_plot = plot_rolling_average_dune( daa_pearl_df, ) with gr.Row(): gr.Markdown(" # Daily active Quickstart agents") with gr.Row(): rolling_avg_plot = plot_rolling_average_dune( daa_qs_df, ) with gr.Row(): gr.Markdown("# 2-weeks rolling average ROI for Pearl agents") with gr.Row(): pearl_rolling_avg_plot = plot_rolling_average_roi( two_weeks_avg_roi_pearl_agents ) 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(): weekly_avg_roi_plot = plot_weekly_average_roi( weekly_avg_roi_df=weekly_avg_roi_pearl_agents, ) with gr.TabItem("🪝 Retention metrics"): with gr.Row(): gr.Markdown("# Wow retention by trader type") with gr.Row(): gr.Markdown( """ Activity based on mech interactions for Olas and non_Olas traders and based on trading acitivity for the unclassified ones. - Olas trader: agent using Mech, with a service ID and the corresponding safe in the registry - Non-Olas trader: agent using Mech, with no service ID - Unclassified trader: agent (safe/EOAs) not using Mechs """ ) with gr.Row(): with gr.Column(scale=1): gr.Markdown("## Wow retention in Pearl markets") wow_retention = calculate_wow_retention_by_type( retention_df, market_creator="pearl" ) wow_retention_plot = plot_wow_retention_by_type( wow_retention=wow_retention ) with gr.Column(scale=1): gr.Markdown("## Wow retention in Quickstart markets") wow_retention = calculate_wow_retention_by_type( retention_df, market_creator="quickstart" ) wow_retention_plot = plot_wow_retention_by_type( wow_retention=wow_retention ) with gr.Row(): gr.Markdown("# Cohort retention graphs") with gr.Row(): gr.Markdown( "The Cohort groups are organized by cohort weeks. A trader is part of a cohort group/week where it was detected the FIRST activity ever of that trader." ) with gr.Row(): gr.Markdown( """ Week 0 for a cohort group is the same cohort week of the FIRST detected activity ever of that trader. Only two values are possible for this Week 0: 1. 100% if the cohort size is > 0, meaning all traders active that first cohort week 2. 0% if the cohort size = 0, meaning no totally new traders started activity that cohort week. """ ) with gr.Row(): with gr.Column(scale=1): gr.Markdown("## Cohort retention of pearl traders") gr.Markdown("### Cohort retention of 🌊 Olas traders") cohort_retention_olas_pearl = calculate_cohort_retention( df=retention_df, market_creator="pearl", trader_type="Olas" ) cohort_retention_plot1 = plot_cohort_retention_heatmap( retention_matrix=cohort_retention_olas_pearl, cmap="Purples" ) with gr.Column(scale=1): gr.Markdown("## Cohort retention of quickstart traders") gr.Markdown("### Cohort retention of 🌊 Olas traders") cohort_retention_olas_qs = calculate_cohort_retention( df=retention_df, market_creator="quickstart", trader_type="Olas" ) cohort_retention_plot4 = plot_cohort_retention_heatmap( retention_matrix=cohort_retention_olas_qs, cmap="Purples", ) with gr.Row(): with gr.Column(scale=1): gr.Markdown("## Cohort retention of pearl traders") cohort_retention_unclassified_pearl = calculate_cohort_retention( df=retention_df, market_creator="pearl", trader_type="unclassified", ) if len(cohort_retention_unclassified_pearl) > 0: gr.Markdown("### Cohort retention of unclassified traders") cohort_retention_plot3 = plot_cohort_retention_heatmap( retention_matrix=cohort_retention_unclassified_pearl, cmap="Greens", ) with gr.Column(scale=1): gr.Markdown("## Cohort retention in quickstart traders") cohort_retention_unclassified_qs = calculate_cohort_retention( df=retention_df, market_creator="quickstart", trader_type="unclassified", ) if len(cohort_retention_unclassified_qs) > 0: gr.Markdown("### Cohort retention of unclassified traders") cohort_retention_plot6 = plot_cohort_retention_heatmap( retention_matrix=cohort_retention_unclassified_qs, cmap="Greens", ) with gr.TabItem("⚙️ Active traders"): with gr.Row(): gr.Markdown("# Active Pearl traders by trader categories") with gr.Row(): active_traders_plot = plot_active_traders(active_traders) with gr.Row(): gr.Markdown("# Active traders for Pearl markets by trader categories") with gr.Row(): active_traders_plot_pearl = plot_active_traders( active_traders, market_creator="pearl" ) with gr.Row(): gr.Markdown("# Active Quickstart traders by trader categories") with gr.Row(): active_traders_plot_qs = plot_active_traders( active_traders, market_creator="quickstart" ) with gr.TabItem("📉 Markets Kullback–Leibler divergence"): with gr.Row(): gr.Markdown( "# Weekly Market Prediction Accuracy for Closed Markets (Kullback-Leibler Divergence)" ) with gr.Row(): gr.Markdown( "Aka, how much off is the market prediction’s accuracy from the real outcome of the event. Values capped at 20 for market outcomes completely opposite to the real outcome." ) with gr.Row(): trade_details_text = get_metrics_text() with gr.Row(): with gr.Column(scale=3): kl_div_plot = plot_kl_div_per_market(closed_markets=closed_markets) with gr.Column(scale=1): interpretation = get_interpretation_text() with gr.TabItem("💰 Money invested per market category"): with gr.Row(): gr.Markdown("# Weekly total bet amount per trader type for all markets") gr.Markdown("## Computed only for traders using the mech service") with gr.Row(): total_bet_amount = plot_total_bet_amount( traders_data, market_filter="all" ) with gr.Row(): gr.Markdown( "# Weekly total bet amount per trader type for Pearl markets" ) with gr.Row(): o_trader_total_bet_amount = plot_total_bet_amount( traders_data, market_filter="pearl" ) with gr.Row(): gr.Markdown( "# Weekly total bet amount per trader type for Quickstart markets" ) with gr.Row(): no_trader_total_bet_amount = plot_total_bet_amount( traders_data, market_filter="quickstart" ) with gr.TabItem("💰 Money invested per single market"): with gr.Row(): gr.Markdown("# Weekly bet amounts per market for all traders") gr.Markdown("## Computed only for traders using the mech service") with gr.Row(): bet_amounts = plot_total_bet_amount_per_trader_per_market(traders_data) with gr.Row(): gr.Markdown("# Weekly bet amounts per market for 🌊 Olas traders") with gr.Row(): o_trader_bet_amounts = plot_total_bet_amount_per_trader_per_market( traders_data, trader_filter="Olas" ) with gr.TabItem("🎖️Weekly winning trades % per trader"): with gr.Row(): gr.Markdown("# Weekly winning trades percentage from all traders") with gr.Row(): metrics_text = get_metrics_text() with gr.Row(): winning_metric = plot_winning_metric_per_trader(weekly_winning_metrics) with gr.Row(): gr.Markdown("# Weekly winning trades percentage from 🌊 Olas traders") with gr.Row(): metrics_text = get_metrics_text() with gr.Row(): winning_metric_olas = plot_winning_metric_per_trader( weekly_winning_metrics_olas ) demo.queue(default_concurrency_limit=40).launch()