Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
Open sidebar
Neel Shah
NIFTy
Commits
3c3518b2
Commit
3c3518b2
authored
Jun 10, 2021
by
Philipp Arras
Browse files
Update description
parent
6b98ae93
Changes
1
Show whitespace changes
Inline
Side-by-side
demos/variational_inference_visualized.py
View file @
3c3518b2
...
@@ -19,13 +19,13 @@
...
@@ -19,13 +19,13 @@
###############################################################################
###############################################################################
# Variational Inference (VI)
# Variational Inference (VI)
#
#
# This script demonstrates how MGVI
and
GeoVI
work for an inference problem
# This script demonstrates how MGVI
,
GeoVI
, MeanfieldVI and FullCovarianceVI
# with only two real quantities of interest. This
enables us to plot the
#
work for an inference problem
with only two real quantities of interest. This
# posterior probability density as two-dimensional plot.
The approximate
#
enables us to plot the
posterior probability density as two-dimensional plot.
# posterior samples are contrasted with the maximum-a-posterior
(MAP) solution
#
The approximate
posterior samples are contrasted with the maximum-a-posterior
# together with samples drawn with the Laplace method. This
method uses the
#
(MAP) solution
together with samples drawn with the Laplace method. This
# local curvature at the MAP solution as inverse covariance of
a Gaussian
#
method uses the
local curvature at the MAP solution as inverse covariance of
# probability density.
#
a Gaussian
probability density.
###############################################################################
###############################################################################
import
numpy
as
np
import
numpy
as
np
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment