%load_ext autoreload
%autoreload 2
from plotlib.main import plot, snsplot, plt, sns
import numpy as np
Call any sns/matplotlib plot throught the plot function
Use the function itself as a kwarg with the parameters to be plotted as a tuple
plot(hist=np.random.randn(1000)); plt.show()
plot(scatter=(np.random.randn(100), np.random.randn(100)))
<matplotlib.collections.PathCollection at 0x7fa83f101610>
Combine Plots
plot(imshow=np.random.randn(500,500)*255,
scatter=(np.random.randn(100)*100, np.random.randn(100)*100))<matplotlib.collections.PathCollection at 0x7fa83f028190>
Use kwargs for more control
plot(hist={'x': np.random.randn(10000), 'bins': 100},
scatter={'x': np.random.randn(80), 'y': 40+10*np.arange(-4, 4, step=0.1)+np.random.randn(80), 'c': 'red'},); plt.show()
plot(boxplot={'x': np.random.randn(1000), 'vert': False});
Same style for sns plotting as well
snsplot(set={'style':"darkgrid"})snsplot(distplot={'a': np.random.randn(1000), 'hist': False, 'kde_kws':{"color": "k", "lw": 3, "label": "KDE"}})<matplotlib.axes._subplots.AxesSubplot at 0x7fa83efccc50>
%%time
fmri = sns.load_dataset("fmri")
snsplot(relplot={'x':"timepoint", 'y':"signal", 'hue':"event", 'style':"event",
'col':"subject", 'col_wrap':5,
'height':3, 'aspect':.75, 'linewidth':2.5,
'kind':"line", 'data':fmri.query("region == 'frontal'")})CPU times: user 2.06 s, sys: 9.42 ms, total: 2.06 s
Wall time: 2.06 s
<seaborn.axisgrid.FacetGrid at 0x7fa83cb0a950>
Easy for automating plots via code
-
iis reserved for axis.
fig, ax = plot(close='all', subplots=(2, 2, {'figsize': (10,10)}))
ax = iter(ax.flat)
font = {'fontsize':10}
plot(
i=next(ax),
set_title=('Number of Videos', font),
set_xlabel=('Number of Videos', font),
set_ylabel=('Number of People', font),
hist=({'x': np.random.randn(1000), 'bins': 0.5+np.arange(-4, 4, 0.1)})
)
plot(
i=next(ax),
set_title=('Number of Videos per Person', font),
set_xlabel=('Number of Videos', font),
set_ylabel=('Number of People', font),
hist=({'x': np.random.randn(1000), 'bins': 28})
)
plot(
i=next(ax),
scatter=({'x': np.random.randn(1000), 'y': np.random.randn(1000)}),
set_title=('Distribution of Video Duration', font),
set_xlabel=('Number of Videos', font),
set_ylabel=('Number of People', font),
hist=({'x': np.random.randn(1000), 'bins': 28})
)
plot(
i=next(ax),
hist = (np.random.randn(1000), {'bins':28}),
set_title = ('Distribution of Average Face Sizes ',font),
**{k:v for k,v in zip(['set_xlabel','set_ylabel'],
[(v,{'fontsize': 10}) for v in
['Average Face Size [pixels]','Number of Videos']])}
); 
