-
Notifications
You must be signed in to change notification settings - Fork 60
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Merge branch 'master' into docs/dtype-object
- Loading branch information
Showing
22 changed files
with
4,955 additions
and
13 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,9 @@ | ||
# Functional API for Optimization | ||
|
||
As an alternative to its object-oriented stateful API, EvoTorch provides an API that conforms to the functional programming paradigm. This functional API has its own advantages like being able to work on not just a single population, but on a batch of populations. | ||
|
||
Here are the examples demonstrating various features of this functional API: | ||
|
||
- **[Maintaining a batch of populations using the functional EvoTorch API](batched_searches.ipynb)**: This notebook shows how one can efficiently run multiple searches simultaneously, each with its own population and hyperparameter configuration, by maintaining a batch of populations. | ||
- **[Solving constrained optimization problems](constrained.ipynb)**: EvoTorch provides batching-friendly constraint penalization functions that can be used with both the object-oriented API and the functional API. In addition, these constraint penalization functions can be used with gradient-based optimization. This notebook demonstrates these features. | ||
- **[Solving reinforcement learning tasks using functional evolutionary algorithms](problem.ipynb)**: The functional evolutionary algorithm implementations of EvoTorch can be used to solve problems that are expressed using the object-oriented core API of EvoTorch. To demonstrate this, this notebook instantiates a `GymNE` problem for the reinforcement learning task "CartPole-v1", and solves it using the functional `pgpe` implementation. |
428 changes: 428 additions & 0 deletions
428
examples/notebooks/Functional_API/batched_searches.ipynb
Large diffs are not rendered by default.
Oops, something went wrong.
Oops, something went wrong.