Evotorch

Latest version: v0.5.1

Safety actively analyzes 682404 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 1 of 2

0.5.1

Fixes
* Fix docs building by Higgcz in https://github.com/nnaisense/evotorch/pull/96

0.5.0

New Features

* Allow the user to reach the search algorithm's internal optimizer by engintoklu in https://github.com/nnaisense/evotorch/pull/89
* Make EvoTorch future-proof by engintoklu in https://github.com/nnaisense/evotorch/pull/77
- Ensure compatibility with PyTorch 2.0 and Brax 0.9
- Migrate from old Gym interface to Gymnasium
* Inform the user when `device` is not set correctly by engintoklu in https://github.com/nnaisense/evotorch/pull/90

Fixes

* Fix `get_minibatch()` of `SupervisedNE` by engintoklu in https://github.com/nnaisense/evotorch/pull/74
* Fix division-by-zero while initializing `CMAES` by engintoklu in https://github.com/nnaisense/evotorch/pull/86
* Fix wrong defaults of `SteadyStateGA` by engintoklu in https://github.com/nnaisense/evotorch/pull/87

0.4.1

Fixes

* Fix the interface of `make_I` (https://github.com/nnaisense/evotorch/pull/62) (engintoklu)
* Fix generate_batch returning `None` (https://github.com/nnaisense/evotorch/pull/63) (engintoklu)
* Fix C decomposition rate calculation on CUDA devices (https://github.com/nnaisense/evotorch/pull/64) (NaturalGradient)

Docs

* Add contribution guidelines (https://github.com/nnaisense/evotorch/pull/70) (engintoklu, Higgcz)
* Add "how to cite" into the README (https://github.com/nnaisense/evotorch/pull/69) (engintoklu)
* Include CMA-ES into the README (https://github.com/nnaisense/evotorch/pull/57) (NaturalGradient)

0.4.0

New Features
* Implementation of `WandbLogger` (https://github.com/nnaisense/evotorch/pull/35) (galatolofederico)
* Simplify the usage of `NeptuneLogger` (https://github.com/nnaisense/evotorch/pull/38) (Higgcz)
* GPU-friendly + vectorized pareto ranking (https://github.com/nnaisense/evotorch/pull/32) (NaturalGradient)
* User interface improvements by (https://github.com/nnaisense/evotorch/pull/34) (engintoklu)
* Add env. variable to control verbosity of the logger (https://github.com/nnaisense/evotorch/pull/48) (Higgcz)
* Add torch-based `CMAES` implementation (https://github.com/nnaisense/evotorch/pull/41) (NaturalGradient)
* Improve `GeneticAlgorithm` and add `MAPElites` (https://github.com/nnaisense/evotorch/pull/44) (engintoklu, pliskowski)
* Add noxfile to run pytest across multiple python versions (https://github.com/nnaisense/evotorch/pull/40) (Higgcz)

Fixes
* Fix all the mkdocstrings warnings (https://github.com/nnaisense/evotorch/pull/39) (Higgcz)
* Fix infinite live reloading of the docs (https://github.com/nnaisense/evotorch/pull/36) (Higgcz)

Docs
* Update the logging page and add `WandbLogger` section (https://github.com/nnaisense/evotorch/pull/37) (Higgcz)

0.3.0

New

**Vectorized gym support**: Added a new problem class, `evotorch.neuroevolution.VecGymNE`, to solve vectorized gym environments. This new problem class can work with brax environments and can exploit GPU acceleration (20).

**PicklingLogger**: Added a new logger, `evotorch.logging.PicklingLogger`, which periodically pickles and saves the current solution to the disk (20).

**Python 3.7 support**: The Python dependency was lowered from 3.8 to 3.7. Therefore, EvoTorch can now be imported from within a Google Colab notebook (16).

API Changes

**pass_info** decorator: When working with `GymNE` (or with the newly introduced `VecGymNE`), if one uses a manual policy class and wishes to receive environment-related information via keyword arguments, that manual policy now needs to be decorated via `pass_info`, as follows: (27)

python
from torch import nn
from evotorch.decorators import pass_info

pass_info
class CustomPolicy(nn.Module):
def __init__(self, **kwargs):
...


**Recurrent policies**: When defining a manual recurrent policy (as a subclass of `torch.nn.Module`) for `GymNE` or for `VecGymNE`, the user is now required to define the forward method of the module according to the following signature:

python
def forward(self, x: torch.Tensor, h: Any = None) -> Tuple[torch.Tensor, Any]:
...


_Note:_ The optional argument `h` is the current state of the network, and the second value of the output tuple is the updated state of the network. A `reset()` method is not required anymore, and it will be ignored (20).

Fixes

Fixed a performance issue caused by the undesired cloning of the entire storages of tensor slices (21).

Fixed the signature and the docstrings of the overridable method `_do_cross_over(...)` of the class `evotorch.operators.CrossOver` (30).

Docs

Added more example scripts and updated the related README file (19).

Updated the documentation related to GPU usage with ray (28).

0.2.0

Fixes:
- Fix docstrings in gaussian.py (11) (engintoklu)
- Fix for str_to_net(...) (12) (engintoklu)
- Hard-code network_device property to CPU for GymNE (6) (NaturalGradient)

Docs:
- Fix comment in the Gym experiments notebook (5) (engintoklu)
- Improve code formatting in docstrings (3) (flukeskywalker)
- Add documentation of NeptuneLogger class (15) (NaturalGradient)

Page 1 of 2

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.