init commit

floodlight/io/sportscode.py

@@ -0,0 +1,65 @@
+import pandas as pd
+from floodlight.core.events import Events
+
+
+def parse_csv_to_event_object(csv_path):
+    """
+    Parse a CSV file into separate event objects based on the 'Timeline'(game) column. Each unique timeline (game) will
+    result in one Event object.
+
+    Parameters
+    ----------
+    csv_path : str
+        Path to the input CSV file. The CSV should contain columns such as 'Row', 'Start time', 'Duration', and 'Timeline'.
+
+    Returns
+    -------
+    list of Events
+        A list containing `Events` objects, one for each unique game (grouped by 'Timeline').
+
+    Notes
+    -----
+    The function performs the following steps:
+    - Reads the CSV file.
+    - Maps column names to internal attributes.
+    - Calculates the 'endtime' as the sum of 'starttime' and 'duration'.
+    - Groups events by 'timeline' to treat each unique timeline as a separate game.
+
+    Example
+    -------
+    csv_path = "path_to_csv"
+    events_objects = parse_csv_to_event_object(csv_path)
+    """
+
+    # Read the CSV file into a pandas DataFrame
+    df = pd.read_csv(csv_path)
+
+    # Create a mapping of column names to their intended event object attributes
+    column_mapping = {
+        "Row": "eID",  # Map the 'Row' column to 'eID'
+        "Start time": "starttime",  # Map 'Start time' to 'starttime'
+        "Duration": "duration",
+        "Timeline": "timeline",
+    }
+
+    # Rename columns based on the mapping
+    df = df.rename(columns=column_mapping)
+
+    # Calculate the endtime
+    df["endtime"] = df["starttime"] + df["duration"]
+
+    # Assign the 'gameclock' column to be the same as 'starttime' (because mandatory for floodlight)
+    df["gameclock"] = df["starttime"]
+
+    # Group by the 'timeline' (which represents a match/game)
+    grouped = df.groupby("timeline")
+
+    events_objects = []
+
+    # Process each group and create an Events object for each game
+    for _, group in grouped:
+        # Create an Events object for the current group (game)
+        events = Events(group)
+        events_objects.append(events)
+
+    return events_objects