Introduction to Distributions¶

When you have a few chains and want to contrast them all with each other, you probably want a summary plot.

To show you how they work, let's make some sample data that all has the same average.

from chainconsumer import Chain, ChainConsumer, Truth, make_sample

# Here's what you might start with
df_1 = make_sample(num_dimensions=4, seed=1, randomise_mean=True)
df_2 = make_sample(num_dimensions=5, seed=2, randomise_mean=True)
print(df_1.head())

Out:

          A         B          C          D  log_posterior
0 -0.458120  4.056907   8.535537  15.745032      -3.709379
1 -0.912067  5.554084   9.716593  14.599073      -2.585703
2  0.266900  4.613901   8.135439  16.295019      -4.087893
3 -0.980402  3.158362  10.061375  13.394896      -5.312100
4 -1.673836  5.256106  10.119388  13.980514      -2.423078

Using distributions¶

# And now we give this to chainconsumer
c = ChainConsumer()
c.add_chain(Chain(samples=df_1, name="An Example Contour"))
fig = c.plotter.plot_distributions()

$A = -0.70^{+0.98}_{-1.04}$, $B = 4.82^{+0.99}_{-1.02}$, $C = 9.38^{+1.03}_{-0.97}$, $D = 14.53^{+1.00}_{-1.02}$

If you want the summary stats you'll need to keep it just one chain. And if you don't want them, you can pass summarise=False to the PlotConfig.

When you add a second chain, you'll see the summaries disappear.

c.add_chain(Chain(samples=df_2, name="Another contour!"))
c.add_truth(Truth(location={"A": 0, "B": 0}))
fig = c.plotter.plot_distributions(col_wrap=3, columns=["A", "B"])