https://pytensor.readthedocs.io/en/latest/index.html 2026-04-28T10:43:41.296835+00:00 ABlog https://pytensor.readthedocs.io/en/latest/index.html/gallery/applications/normalizing_flows_in_pytensor.html 2025-08-17T00:00:00+00:00 Ricardo Vieira <p class="ablog-post-excerpt"><p>A hip new algorithm for doing machine learning on distributions is called <em>normalizing flows</em>.</p> </p>

A hip new algorithm for doing machine learning on distributions is called normalizing flows.

2025-08-17T00:00:00+00:00 https://pytensor.readthedocs.io/en/latest/index.html/gallery/introduction/pytensor_intro.html 2025-08-16T00:00:00+00:00 Ricardo Vieira <p class="ablog-post-excerpt"><p>A library to define, manipulate, and compile computational graphs.</p> </p>

A library to define, manipulate, and compile computational graphs.

2025-08-16T00:00:00+00:00 https://pytensor.readthedocs.io/en/latest/index.html/gallery/optimize/root.html 2025-06-12T00:00:00+00:00 Jesse Grabowski <p class="ablog-post-excerpt"><p>When faced with problems involving systems of nonlinear equations, it is rare to actually have access to analytic solutions for the zeros of the system. Nevertheless, these zeros are often important to downstream tasks. A common application is in perturbation theory, where we seek to linearize a nonlinear system around the fixed points of that system.</p> </p>

When faced with problems involving systems of nonlinear equations, it is rare to actually have access to analytic solutions for the zeros of the system. Nevertheless, these zeros are often important to downstream tasks. A common application is in perturbation theory, where we seek to linearize a nonlinear system around the fixed points of that system.

2025-06-12T00:00:00+00:00 https://pytensor.readthedocs.io/en/latest/index.html/gallery/rewrites/graph_rewrites.html 2025-01-11T00:00:00+00:00 Ricardo Vieira <p class="ablog-post-excerpt"><p>This section walks through the low level details of PyTensor graph manipulation. Users are not supposed to work or even be aware of these details, but it may be helpful for developers. We start with very <strong>bad practices</strong> and move on towards the <strong>right</strong> way of doing rewrites.</p> </p>

This section walks through the low level details of PyTensor graph manipulation. Users are not supposed to work or even be aware of these details, but it may be helpful for developers. We start with very bad practices and move on towards the right way of doing rewrites.

2025-01-11T00:00:00+00:00 https://pytensor.readthedocs.io/en/latest/index.html/gallery/scan/scan_tutorial.html 2025-01-11T00:00:00+00:00 Jesse Grabowski <p class="ablog-post-excerpt"><p>A Pytensor function graph is composed of two types of nodes: Variable nodes which represent data, and Apply node which apply Ops (which represent some computation) to Variables to produce new Variables.</p> </p>

A Pytensor function graph is composed of two types of nodes: Variable nodes which represent data, and Apply node which apply Ops (which represent some computation) to Variables to produce new Variables.

2025-01-11T00:00:00+00:00