|
import pandas as pd |
|
|
|
def create_player_exposures(df: pd.DataFrame, player_columns: list, entrants: list = None): |
|
player_frame = pd.DataFrame() |
|
if entrants is not None: |
|
overall_players = pd.Series(list(df[df['BaseName'].isin(entrants)][player_columns].values.flatten())).value_counts() |
|
else: |
|
overall_players = pd.Series(list(df[player_columns].values.flatten())).value_counts() |
|
top_1per_players = pd.Series(list(df[df['percentile_finish'] <= 0.01][player_columns].values.flatten())).value_counts() |
|
top_5per_players = pd.Series(list(df[df['percentile_finish'] <= 0.05][player_columns].values.flatten())).value_counts() |
|
top_10per_players = pd.Series(list(df[df['percentile_finish'] <= 0.10][player_columns].values.flatten())).value_counts() |
|
top_20per_players = pd.Series(list(df[df['percentile_finish'] <= 0.20][player_columns].values.flatten())).value_counts() |
|
contest_len = len(df) |
|
len_1per = len(df[df['percentile_finish'] <= 0.01]) |
|
len_5per = len(df[df['percentile_finish'] <= 0.05]) |
|
len_10per = len(df[df['percentile_finish'] <= 0.10]) |
|
len_20per = len(df[df['percentile_finish'] <= 0.20]) |
|
each_set_name = ['Overall', ' Top 1%', ' Top 5%', 'Top 10%', 'Top 20%'] |
|
each_frame_set = [overall_players, top_1per_players, top_5per_players, top_10per_players, top_20per_players] |
|
each_len_set = [contest_len, len_1per, len_5per, len_10per, len_20per] |
|
player_count_var = 0 |
|
for each_set in each_frame_set: |
|
set_frame = each_set.to_frame().reset_index().rename(columns={'index': 'Player', 'count': 'Count'}) |
|
set_frame['Percent'] = set_frame['Count'] / each_len_set[player_count_var] |
|
set_frame = set_frame[['Player', 'Percent']] |
|
set_frame = set_frame.rename(columns={'Percent': f'Exposure {each_set_name[player_count_var]}'}) |
|
if len(player_frame) == 0: |
|
player_frame = set_frame |
|
else: |
|
player_frame = pd.merge(player_frame, set_frame, on='Player', how='outer') |
|
player_count_var += 1 |
|
|
|
return player_frame |