Back to Cheatsheets

Table of Contents

1. ABOUT

• Import
• Show
• Setup

2. FORMATTING

• Labels
• Modify Ticks
• Style
• Despine
• Scale
• Palettes

3. BARPLOT

• Error Bars
• Aggregate
• Aggregate by multiple

4. DISTRIBUTIONS

• KDE plots
• Box plots
• Violin plots

ABOUT

• based on Matplotlib, but improves
• provides a more visually appealing plotting style and concise syntax.
• natively understands Pandas DataFrames, making it easier to plot data directly from CSVs.
• easily summarize Pandas DataFrames with many rows of data into aggregated charts.

Import

import seaborn as sns

Show

plt.show()

Setup

f, ax = plt.subplots(figsize=(width,height))

FORMATTING

Labels

ax.set_title() : same as plt.title()

Modify Ticks

import matplotlib.ticker as mtick
fmt = '${x:,.0f}'
tick = mtick.StrMethodFormatter(fmt)
ax.yaxis.set_major_formatter(tick)

• add $ to tick names on y-axis

Style

sns.set_style('style')` 

• default is ‘darkgrid’
• built in themes : darkgrid, whitegrid, dark, white, ticks
• background colours : dark, white, tick
• grids: whitegrid, white(no grid)

Despine

sns.despine()

• change usage of spines i,e, the borders of the figure
• removes top and right spines
• (left=True) : pass in to also remove other spinrs
• normal default without this call is four spines
• must call have sns.plot_func()

Scale

sns.set_context('scale_type', font_scale=text_size, rc={"grid.linewidth": 5})

• scale_type : presets = paper, notebook(default), talk, poster
• rc: run command, override one of the commands called with scale_type
  • poster =

    {'axes.labelsize': 17.6,
     'axes.titlesize': 19.200000000000003,
     'font.size': 19.200000000000003,
     'grid.linewidth': 1.6,
     'legend.fontsize': 16.0,
     'lines.linewidth': 2.8000000000000003,
     'lines.markeredgewidth': 0.0,
     'lines.markersize': 11.200000000000001,
     'patch.linewidth': 0.48,
     'xtick.labelsize': 16.0,
     'xtick.major.pad': 11.200000000000001,
     'xtick.major.width': 1.6,
     'xtick.minor.width': 0.8,
     'ytick.labelsize': 16.0,
     'ytick.major.pad': 11.200000000000001,
     'ytick.major.width': 1.6,
     'ytick.minor.width': 0.8}
    

Palettes

sns.set_palette('palette_name')

• if not called will use default colours
• defaults: depp, muted, pastel, bright, dark, colorblind
• sns.color_palettte(list_of_colours): create own pallette and then call set, or pass in name of a color brewer palette and # of variables
• sns.palplot(palette_var_name) : visualise colours of palette

BARPLOT

sns.barplot(data=df, x='x_col_name', y='y_col_name')

• by defalt will aggregate y and plot mean of each x
• sam in matplotlib as subplot() + bar() + xticks() + xticklabels() +xlabel() + ylabel()

Error Bars

sns.barplot(...., ci='error_measure')

• adds error bars by default using bootstrapped confidence interval (95%)
• ci='sd' : use standard deviation for error bar instead

Aggregate

sns.barplot(...., estimator=aggregate_func)

• np.median : many outliers
• len : categorical and want to count how many times each category appears (eg. survey respones)
• by default calculates mean

Aggregate by multiple

hue='other_x_col_name'

• nested categorical value, colour data based on 3rd variable

DISTRIBUTIIONS

• univariate: only has one variable, one-dimensional
  • KDE plots
  • Box plots

KDE plots

sns.kdeplot(dataset1, shade=True)
sns.kdeplot(dataset2, shade=True)

• Kernel Density Estimator
• sense of a univariate as a curve, shows shape of the data
• preferable to histogra, smooth the datasets, allow to generalise over the shape of our data (not specific data points)
• shade : when area under the curve is shaded
• data : pandas dataframe, python list, numpy array

Box plots

sns.boxplot(data=df, x='df_x_col', y='df_y_col')

• shows range of dataset eg. outliers, where significant portion of data lies
• The box represents the interquartile range
• The line in the middle of the box is the median
• The end lines are the first and third quartiles
• The diamonds show outliers
• x and y are one dimensional set of values

Violin plots

sns.violinplot(data=df, x='df_x_col', y='df_y_col, hue='3rd_variable', split=True)

• good for showing multiple distributions at 1 time
• maps each individual data point, get an estimation of the dataset thanks to KDE
• two KDE plots that are symmetrical along the center line
• white dot represent the median
• thick black line in the center of each violin represents the interquartile range
• lines that extend form the center are the confidence intervals (95%)
• split shows the hue together instead of side by side

STRIP PLOTS

sns.stripplot(x='x', y='y', data=df)