fix: clean up seaborn chapter to avoid warnings

README.md

@@ -23,6 +23,9 @@ If I were using Windows, it might look like this:
 Set these aside for now and we'll pick them up in chapter 2.
 
 ## Changelog
+### v0.2.0 (2024-06-18)
+Cleaned up plotting chapter to avoid misc Seaborn warnings.
+
 ### v0.1.1 (2024-04-22)
 Fix Pandas `mean` example. More: Pandas changed their defaults to throw an
 error if you try to call this on string columns. Fixed example to explicitely

code/06_01_summary.py

@@ -6,7 +6,6 @@
 from os import path
 import matplotlib.image as mpimg
 
-pd.options.mode.chained_assignment = None
 %matplotlib qt
 
 # change this to the directory where the csv files that come with the book are
@@ -16,11 +15,6 @@
 DATA_DIR = './data'
 FIG_DIR = './figures'
 
-dftm = dfpm.groupby(['match_id', 'team']).agg(
-    total_shot=('shot', 'sum'),
-    total_pass=('pass', 'sum'),
-    total_goal=('goal', 'sum'))
-
 ###############
 # distributions
 ###############
@@ -34,7 +28,7 @@
 
 # processing
 dfpm = pd.merge(dfpm, dfm[['match_id', 'home_team', 'away_team']], how='left')
-dfpm['opp'] = np.nan
+dfpm['opp'] = pd.NA
 dfpm.loc[dfpm['team'] == dfpm['home_team'], 'opp'] = dfpm['away_team']
 dfpm.loc[dfpm['team'] == dfpm['away_team'], 'opp'] = dfpm['home_team']
 
@@ -60,43 +54,39 @@
 ##########
 
 # basic displot - all on one line
-g = (sns.FacetGrid(dfpm).map(sns.kdeplot, 'pass', shade=True))
+g = (sns.FacetGrid(dfpm).map(sns.kdeplot, 'pass', fill=True))
 g.set(xlim=(-5, 120))
-plt.show()
 
 # on seperate lines so it's clearer it's a two step process
 g = (sns.FacetGrid(dfpm)
-     .map(sns.kdeplot, 'dist', shade=True))
+     .map(sns.kdeplot, 'dist', fill=True))
 
 # hue
 g = (sns.FacetGrid(dfpm, hue='pos')
-     .map(sns.kdeplot, 'pass', shade=True)
+     .map(sns.kdeplot, 'pass', fill=True)
      .add_legend()
      .set(xlim=(-5, 120)))
-plt.show()
 
 # add col
 g = (sns.FacetGrid(dfpm, hue='pos', col='side')
-     .map(sns.kdeplot, 'pass', shade=True)
+     .map(sns.kdeplot, 'pass', fill=True)
      .add_legend()
      .set(xlim=(-5, 120)))
-plt.show()
 
 # add col order
 g = (sns.FacetGrid(dfpm, hue='pos', col='side', col_order=['left', 'central',
                        'right'])
-     .map(sns.kdeplot, 'pass', shade=True)
+     .map(sns.kdeplot, 'pass', fill=True)
      .add_legend()
      .set(xlim=(-5, 160)))
-plt.show()
 
 # rows
 dfpm.loc[dfpm['pos'] == 'GKP', 'side'] = 'central'
 g = (sns.FacetGrid(dfpm, hue='pos', col='side', row='pos',
                    col_order=['left', 'central', 'right'],
                    row_order=['FWD', 'MID', 'DEF', 'GKP'],
                    )
-     .map(sns.kdeplot, 'pass', shade=True)
+     .map(sns.kdeplot, 'pass', fill=True)
      .add_legend()
      .set(xlim=(-5, 160)))
 
@@ -116,7 +106,7 @@
  .sort_values('date').head())
 
 def home_away_score_df(df, location):
-    df = df[['match_id', 'date', f'{location}_team', f'{location}_score']]
+    df = df[['match_id', 'date', f'{location}_team', f'{location}_score']].copy()
     df.columns = ['match_id', 'date', 'team', 'score']
     df['location'] = location
     return df
@@ -129,9 +119,8 @@ def home_away_score_df(df, location):
 
 # now can plot points by scoring system and position
 g = (sns.FacetGrid(score_long, hue='location')
-     .map(sns.kdeplot, 'score', shade=True))
+     .map(sns.kdeplot, 'score', fill=True))
 g.add_legend()
-plt.show()
 
 #################################
 # relationships between variables
@@ -141,7 +130,7 @@ def home_away_score_df(df, location):
 g = sns.relplot(x='weight', y='height', data=dfp)
 
 # with hue
-g = sns.relplot(x='jweight', y='jheight', hue='pos', data=dfp)
+g = sns.relplot(x='weight', y='height', hue='pos', data=dfp)
 
 # adding jitter
 import random
@@ -161,15 +150,15 @@ def home_away_score_df(df, location):
 # and label_order arguments
 g = (sns.FacetGrid(dfp, hue='team', col='grouping', col_wrap=2, aspect=2,
                    col_order=['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H'])
-    .map(sns.kdeplot, 'weight', shade=True)
+    .map(sns.kdeplot, 'weight', fill=True)
     .add_legend())
 
 # contour plots
-# shade
+# fill
 g = (sns.FacetGrid(dfp, col='pos', hue='pos', col_wrap=2, aspect=2)
-     .map(sns.kdeplot, 'weight', 'height', shade=True))
+     .map(sns.kdeplot, 'weight', 'height', fill=True))
 
-# no shade
+# no fill
 g = (sns.FacetGrid(dfp, col='pos', hue='pos', col_wrap=2, aspect=2)
      .map(sns.kdeplot, 'weight', 'height'))
 
@@ -215,25 +204,22 @@ def home_away_score_df(df, location):
 g = sns.relplot(x='min_period', y='dist_m', kind='line', hue='foot',
                 data=dfs.query("period in ('1H', '2H')"),
                 row='period')
-plt.show()
 
 ##############
 # plot options
 ##############
 
 # basic plot
 g = (sns.FacetGrid(dfpm, col='pos')
-     .map(sns.kdeplot, 'pass', shade=True))
-
-plt.show()
+     .map(sns.kdeplot, 'pass', fill=True))
 
 # wrap columns
 g = (sns.FacetGrid(dfpm, col='pos', col_wrap=2)
-     .map(sns.kdeplot, 'pass', shade=True))
+     .map(sns.kdeplot, 'pass', fill=True))
 
 # adding a title
-g.fig.subplots_adjust(top=0.9)
-g.fig.suptitle('Distribution of No of Passes by Position')
+g.figure.subplots_adjust(top=0.9)
+g.figure.suptitle('Distribution of No of Passes by Position')
 
 # modifying options
 g.set(xlim=(-5, 120))
@@ -243,75 +229,3 @@ def home_away_score_df(df, location):
 
 # saving
 g.savefig(path.join(FIG_DIR, 'no_passes_by_position.png'))
-
-#############
-# shot charts
-#############
-
-dfs[['name', 'dist', 'foot', 'goal', 'x1', 'y1']].head(5)
-
-dfs['x'] = dfs['x1']*120/100
-dfs['y'] = (100 - dfs['y1'])*75/100
-
-# shot data
-g = sns.relplot(data=dfs, x='x', y='y', kind='scatter')
-g.set(yticks=[], xticks=[], xlabel=None, ylabel=None)
-g.despine(left=True, bottom=True)
-
-map_img = mpimg.imread('./fig/soccer_field.png')
-
-# scatter plot with field overlay
-g = sns.relplot(data=dfs, x='x', y='y', kind='scatter')
-g.set(yticks=[], xticks=[], xlabel=None, ylabel=None)
-g.despine(left=True, bottom=True)
-for ax in g.fig.axes:
-    ax.imshow(map_img, zorder=0, extent=[0, 120, 0, 75])
-
-plt.show()
-
-# how about a bit of jitter
-dfs['xj'] = dfs['x'].apply(lambda x: x + random.gauss(0, 1))
-dfs['yj'] = dfs['y'].apply(lambda x: x + random.gauss(0, 1))
-
-g = sns.relplot(data=dfs, x='xj', y='yj', kind='scatter')
-for ax in g.fig.axes:
-    ax.imshow(map_img, zorder=0, extent=[0, 115, 0, 74])
-g.set(yticks=[], xticks=[], xlabel=None, ylabel=None)
-g.despine(left=True, bottom=True)
-plt.show()
-
-# putting it in a function
-def shot_chart(df, **kwargs):
-    g = sns.relplot(data=df, x='xj', y='yj', kind='scatter', **kwargs)
-    g.set(yticks=[], xticks=[], xlabel=None, ylabel=None)
-    g.despine(left=True, bottom=True)
-
-    for ax in g.fig.axes:
-        ax.imshow(map_img, zorder=0, extent=[0, 115, 0, 74])
-
-    return g
-
-shot_chart(dfs, hue='goal', style='goal')
-shot_chart(dfs, col='foot', hue='foot')
-
-# and columns
-shot_chart(dfs, hue='goal', style='goal', col='team', height=3, col_wrap=4,
-           s=10)
-
-# now let's try a contour plot
-g = (sns.FacetGrid(dfs, col='foot', hue='foot')
-     .map(sns.kdeplot, 'x', 'y', alpha=0.5, shade=True)
-     .add_legend())
-g.set(yticks=[], xticks=[], xlabel=None, ylabel=None)
-g.despine(left=True, bottom=True)
-for ax in g.fig.axes:
-    ax.imshow(map_img, zorder=0, extent=[0, 115, 0, 74])
-
-# add goal row, turn shading off
-g = (sns.FacetGrid(dfs, col='foot', hue='foot')
-     .map(sns.kdeplot, 'x', 'y', alpha=0.5)
-     .add_legend())
-g.set(yticks=[], xticks=[], xlabel=None, ylabel=None)
-g.despine(left=True, bottom=True)
-for ax in g.fig.axes:
-    ax.imshow(map_img, zorder=0, extent=[0, 115, 0, 74])

code/06_02_shot_chart.py

@@ -1,12 +1,9 @@
 import pandas as pd
 import random
-from pandas import DataFrame
 import seaborn as sns
-import numpy as np
 import matplotlib.pyplot as plt
 from os import path
 
-pd.options.mode.chained_assignment = None
 %matplotlib qt
 
 DATA_DIR = './data'
@@ -38,7 +35,7 @@
 g.despine(left=True, bottom=True)
 
 import matplotlib.image as mpimg
-map_img = mpimg.imread('./fig/soccer_field.png')
+map_img = mpimg.imread('./data/soccer_field.png')
 
 # scatter plot with field overlay
 g = sns.relplot(data=dfs, x='x', y='y', kind='scatter', size=5)
@@ -93,13 +90,13 @@ def add2_flexible(num1, num2, **kwargs):
      .map(sns.kdeplot, 'x', 'y', alpha=0.5))
 g.set(yticks=[], xticks=[], xlabel=None, ylabel=None)
 g.despine(left=True, bottom=True)
-for ax in g.fig.axes:
+for ax in g.figure.axes:
     ax.imshow(map_img, zorder=0, extent=[0, 120, 0, 75])
 
 # turn shading off - by team
 g = (sns.FacetGrid(dfs, col='team', col_wrap=4, height=2, hue='team')
      .map(sns.kdeplot, 'x', 'y', alpha=0.5))
 g.set(yticks=[], xticks=[], xlabel=None, ylabel=None)
 g.despine(left=True, bottom=True)
-for ax in g.fig.axes:
+for ax in g.figure.axes:
     ax.imshow(map_img, zorder=0, extent=[0, 120, 0, 75])

solutions-to-exercises/06_plotting_answers.py

@@ -17,9 +17,9 @@
 
 # 6.1a
 g = (sns.FacetGrid(dftm)
-     .map(sns.kdeplot, 'pass', shade=True))
-g.fig.subplots_adjust(top=0.9)
-g.fig.suptitle('Distribution of Passes')
+     .map(sns.kdeplot, 'pass', fill=True))
+g.figure.subplots_adjust(top=0.9)
+g.figure.suptitle('Distribution of Passes')
 g.savefig('./solutions-to-exercises/6-1a.png')
 
 # Now modify your plot to show the distribution of passes by whether the team
@@ -28,41 +28,41 @@
 
 # 6.1b
 g = (sns.FacetGrid(dftm, hue='win')
-    .map(sns.kdeplot, 'pass', shade=True))
-g.fig.subplots_adjust(top=0.9)
-g.fig.suptitle('Distribution of Passes by Win/Loss B')
+    .map(sns.kdeplot, 'pass', fill=True))
+g.figure.subplots_adjust(top=0.9)
+g.figure.suptitle('Distribution of Passes by Win/Loss B')
 g.savefig('./solutions-to-exercises/6-1b.png')
 
 # 6.1c
 g = (sns.FacetGrid(dftm, col='win')
-    .map(sns.kdeplot, 'pass', shade=True))
-g.fig.subplots_adjust(top=0.8)
-g.fig.suptitle('Distribution of Passes by Win/Loss C')
+    .map(sns.kdeplot, 'pass', fill=True))
+g.figure.subplots_adjust(top=0.8)
+g.figure.suptitle('Distribution of Passes by Win/Loss C')
 g.savefig('./solutions-to-exercises/6-1c.png')
 
 # 6.1d
 
 g = (sns.FacetGrid(dftm, col='win', hue='win')
-    .map(sns.kdeplot, 'pass', shade=True))
-g.fig.subplots_adjust(top=0.8)
-g.fig.suptitle('Distribution of Passes by Win/Loss D')
+    .map(sns.kdeplot, 'pass', fill=True))
+g.figure.subplots_adjust(top=0.8)
+g.figure.suptitle('Distribution of Passes by Win/Loss D')
 g.savefig('./solutions-to-exercises/6-1d.png')
 
 
 # 6.1e
 g = (sns.FacetGrid(dftm, col='team', col_wrap=6)
-    .map(sns.kdeplot, 'pass', shade=True))
-g.fig.subplots_adjust(top=0.9)
-g.fig.suptitle('Distribution of Passes by Team')
+    .map(sns.kdeplot, 'pass', fill=True))
+g.figure.subplots_adjust(top=0.9)
+g.figure.suptitle('Distribution of Passes by Team')
 g.savefig('./solutions-to-exercises/6-1e.png')
 
 ###############################################################################
 # 6.2
 ###############################################################################
 # 6.2a 
 g = sns.relplot(x='pass', y='pass_opp', data=dftm)
-g.fig.subplots_adjust(top=0.9)
-g.fig.suptitle('Passes vs Opponent Passes')
+g.figure.subplots_adjust(top=0.9)
+g.figure.suptitle('Passes vs Opponent Passes')
 g.savefig('./solutions-to-exercises/6-2a.png')
 
 # 6.2b