Robust statistics linear regression in seaborn pairplot - statistics

Question

Trying to implement robust statistics instead of ordinary least squares (OLS) fitting so that outliers aren't such a problem to my fits.
I was hoping to implement this in the pairplot function of seaborn and can't see and easy way to add this from the AP documentation as there doesn't seem to be a key word argument for the fit.
From: scipy lectures They suggest using the following but I guess thats for regplot where you can define the fit using
`fit = statsmodels.formula.api.rlm()`
Here is some sample code
import seaborn as sns; sns.set(style="ticks", color_codes=True)
import matplotlib.pyplot as plt
%matplotlib inline
iris = sns.load_dataset("iris")
sns.pairplot(iris, kind="reg")#, robust = True)
plt.show()
Thanks in advance!
Edit: I found a workaround, but loose the 'hue' function apparently that could be done on the pairplot. Would be a nice feature to add robust option to pairplot.
Code:
def corrfunc(x, y, **kws):
r, _ = stats.pearsonr(x, y)
ax = plt.gca()
ax.annotate("r = {:.2f}".format(r), xy=(.1, .9), xycoords=ax.transAxes)
g = sns.PairGrid(df1, palette=["red"])
g.map_upper(sns.regplot, robust = True)
g.map_diag(sns.distplot, kde=True)
g.map_lower(sns.kdeplot, cmap="Blues_d")
g.map_lower(corrfunc)

Answer 1

Extra keywords, such as "robust = True" can be passed to regplot via the plot_kws argument:
sns.pairplot(df1,kind='reg',hue='species',plot_kws=dict(robust=True,n_boot=50))
NB: In this example I have also decreased n_boot to reduce the runtime (see "robust" in regplot documentation), so the confidence intervals might be incorrect).