updating data and graphs

app.py

@@ -13,6 +13,7 @@ from tabs.dist_gap import (
     get_distribution_plot,
     get_correlation_map,
     get_kde_with_trades,
+    get_regplot_with_mean_trade_size,
 )
 
 
@@ -130,7 +131,15 @@ with demo:
                 kde_trades_plot = get_kde_with_trades(markets_data)
 
             with gr.Row():
-                gr.Markdown("# Correlation analysis between variables")
+                gr.Markdown("# Relationship with mean trade size")
+
+            with gr.Row():
+                reg_plot = get_regplot_with_mean_trade_size(markets_data)
+
+            with gr.Row():
+                gr.Markdown(
+                    "# Correlation analysis between the metric and market variables"
+                )
 
             with gr.Row():
                 correlation_plot = get_correlation_map(markets_data)

live_data/markets_live_data.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f1aefc2dd441883aca8a95db7715a511b763a5b486307a903dcea30df7ef828
-size 27422
+oid sha256:1e0451385c1de2891fe1ab7fff8b3e5a5f797f351b2ca799266e65adec57a9b9
+size 34357

live_data/markets_live_data_sample.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e601332794f53f1c65384434aa7bbcad617853f3aa7f89eeb68640f36edc7b14
-size 22201
+oid sha256:8592b33790b8209ebd8de3ac3108e76f25a3e02fc5720dc0a8984fe47e09cad1
+size 27990

notebooks/market_trades.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7466aaeee89b4eb10c8b3902231126fc480c8829e792071784b7b38be56a9208
+size 20405

scripts/live_markets_data.py

@@ -231,6 +231,7 @@ def transform_fpmms(fpmms: pd.DataFrame, filename: str, current_timestamp: int)
     # computation of token distributions
     fpmms["token_first_amount"] = fpmms.outcomeTokenAmounts.apply(lambda x: int(x[0]))
     fpmms["token_second_amount"] = fpmms.outcomeTokenAmounts.apply(lambda x: int(x[1]))
+    fpmms["liquidityMeasure"] = fpmms["liquidityMeasure"].apply(lambda x: int(x))
     fpmms["total_tokens"] = fpmms.apply(
         lambda x: x.token_first_amount + x.token_second_amount, axis=1
     )

scripts/live_traders_data.py

@@ -12,8 +12,9 @@ from utils import (
 )
 from utils import SUBGRAPH_API_KEY, _to_content
 from queries import omen_trader_votes_query
+from live_markets_data import logger
 
-logger = logging.getLogger(__name__)
+#logger = logging.getLogger(__name__)
 
 headers = {
     "Accept": "application/json, multipart/mixed",
@@ -90,25 +91,6 @@ def transform_trades(trades_json: dict) -> pd.DataFrame:
     return df
 
 
-def compute_from_timestamp_value(
-    fpmm_id: str, opening_timestamp: int, fpmms: pd.DataFrame
-) -> Optional[int]:
-    """Function to find the latest timestamp registered for a specific market"""
-    try:
-        market_data = fpmms.loc[fpmms["id"] == fpmm_id]
-        # how many previous samples do we have?
-        if len(market_data) == 1:
-            # take the opening Timestamp of the Market
-            return opening_timestamp
-        timestamps = (market_data.tokens_timestamp.values).sort()
-        # the last value is the current timestamp so we need to take the previous one
-        return timestamps[-2]
-    except Exception as e:
-        logger.error(
-            f"Error when trying to get the from timestamp value of the market id {fpmm_id}"
-        )
-        return None
-
 
 def compute_votes_distribution(market_trades: pd.DataFrame):
     """Function to compute the distribution of votes for the trades of a market"""
@@ -129,8 +111,8 @@ def add_trading_info(fpmms: pd.DataFrame, current_timestamp: int) -> None:
     # Iterate over the markets
     for i, fpmm in tqdm(fpmms.iterrows(), total=len(fpmms), desc="Analysing trades"):
         # update the trades for this market and at this specific current_timestamp
-        logger.debug(f"current timestamp = {current_timestamp} and market timestamp={fpmm["tokens_timestamp"]}")
-        to_update = fpmm["open"] and fpmm["tokens_timestamp"] == current_timestamp
+        logger.debug(f"current timestamp = {current_timestamp} and market timestamp={fpmm["sample_timestamp"]}")
+        to_update = fpmm["open"] and fpmm["sample_timestamp"] == current_timestamp
         if not to_update:  # jump closed markets or old data
             logger.debug("Jumping this row")
             continue
@@ -141,12 +123,18 @@ def add_trading_info(fpmms: pd.DataFrame, current_timestamp: int) -> None:
             fpmm_id=market_id,
         )
         market_trades = transform_trades(market_trades_json)
-        if len(market_trades) == 0:
-            logger.info("No trades for this market")
-            continue
-
         fpmms.at[i,"total_trades"] = len(market_trades)
-        first_outcome, second_outcome = compute_votes_distribution(market_trades)
+
+        if len(market_trades) > 0:
+            # adding average trade size
+            market_trades["collateralAmountUSD"] = market_trades.collateralAmountUSD.apply(lambda x: round(float(x),3))
+            mean_trade_size = market_trades.collateralAmountUSD.mean()
+            first_outcome, second_outcome = compute_votes_distribution(market_trades)
+        else:
+            logger.info("No trades for this market")
+            mean_trade_size = 0.0
+            first_outcome, second_outcome = 50.0, 50.0
+        fpmms.at[i,"mean_trade_size"] = mean_trade_size
         logger.info(
             f"first outcome votes ={first_outcome}, second outcome votes = {second_outcome}"
         )

tabs/dist_gap.py

@@ -36,18 +36,27 @@ def get_kde_with_trades(markets_data: pd.DataFrame):
     return gr.Plot(value=plot.get_figure())
 
 
+def get_regplot_with_mean_trade_size(markets_data: pd.DataFrame):
+    """Function to Plot data and a linear regression model fit"""
+    regplot = sns.regplot(live_fpmms, x="dist_gap_perc", y="mean_trade_size")
+    plt.ylabel("Mean trade size in USD")
+    return gr.Plot(value=regplot.get_figure())
+
+
 def get_correlation_map(markets_data: pd.DataFrame):
     """Function to paint the correlation between different variables"""
 
-    columns_of_interest = ["total_trades", "dist_gap_perc", "days_to_resolution"]
+    columns_of_interest = [
+        "total_trades",
+        "dist_gap_perc",
+        "liquidityMeasure",
+        "mean_trade_size",
+    ]
     data = markets_data[columns_of_interest]
 
     # Compute the correlation matrix
     correlation_matrix = data.corr()
 
-    # Create a figure and axis
-    # plt.figure(figsize=(10, 8))
-
     # Create the heatmap
     heatmap = sns.heatmap(
         correlation_matrix,