Dynamic Computation Graphs
Explore how PyTorch builds dynamic computation graphs for automatic differentiation. Understand the visualization of tensors and gradients, and see how control flow can shape complex graphs in your models.
Introduction to dynamic computation graph
“Unfortunately, no one can be told what the dynamic computation graph is. You have to see it for yourself."
- Morpheus
How great was The Matrix? Right? Right? But jokes aside, we want you to see the graph for yourself too!
The PyTorchViz package and its make_dot(variable) method allow us to easily visualize a graph associated with a given Python variable involved in the gradient ...