def test_distplot(n=100):
a = [np.random.randn(n) for i in range(10)]
x = range(len(a))
ax = distplot(x, a, showoutliers=1)
ax = distplot(x, a, color='r')
ax = distplot(x, a, label='points', color='b', showmean=1)
ax.legend()
def test_distplot(n=100):
a = [np.random.randn(n) for i in range(10)]
x = range(len(a))
ax = distplot(x, a, showoutliers=1)
ax = distplot(x, a, color="r")
ax = distplot(x, a, label="points", color="b", showmean=1)
ax.legend()
def plot_autocorr(chain):
""" Plot the autocorrelation of parameters in the chain. This can
be slow if there are many parameters.
"""
nwalkers, nsamples, npar = chain.shape
nrows, ncols = get_nrows_ncols(npar)
fig,axes = get_fig_axes(nrows, ncols, npar)
for i,ax in enumerate(axes):
acor = [autocorr(chain[j,:,i], maxlag=150) for j in xrange(nwalkers)]
distplot(np.transpose(acor), ax=ax)
ax.axhline(0, color='r', lw=0.5)
puttext(0.1, 0.1, P['names'][i], ax, fontsize=16)
return fig, axes
def plot_autocorr(chain):
""" Plot the autocorrelation of parameters in the chain. This can
be slow if there are many parameters.
"""
nwalkers, nsamples, npar = chain.shape
nrows, ncols = get_nrows_ncols(npar)
fig, axes = get_fig_axes(nrows, ncols, npar)
for i, ax in enumerate(axes):
acor = [autocorr(chain[j, :, i], maxlag=150) for j in xrange(nwalkers)]
distplot(np.transpose(acor), ax=ax)
ax.axhline(0, color='r', lw=0.5)
puttext(0.1, 0.1, P['names'][i], ax, fontsize=16)
return fig, axes
