https://github.com/mynameisvinn/dawkins
evolution strategies for classification
https://github.com/mynameisvinn/dawkins
evolution-strategies machine-learning optimization
Last synced: 8 months ago
JSON representation
evolution strategies for classification
- Host: GitHub
- URL: https://github.com/mynameisvinn/dawkins
- Owner: mynameisvinn
- Created: 2018-05-09T14:09:37.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2018-05-11T15:01:21.000Z (over 7 years ago)
- Last Synced: 2024-12-27T14:25:45.093Z (10 months ago)
- Topics: evolution-strategies, machine-learning, optimization
- Language: Python
- Homepage:
- Size: 15.6 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# dawkins
apply evolution strategies to supervised, classification tasks. at every iteration (“generation”), a population of parameter vectors (“genotypes”) is perturbed (“mutated”) and their objective function value (“fitness”) is evaluated.
## why evolution strategies (es)?
es is an optimization technique that learns parameters without backpropagation. no gradients are computed - why do things the easy way when you can do it the hard way?
## example
if you know scikit, you know the drill.
```python
# iris dataset
iris = learn.datasets.load_dataset('iris')
X = iris.data
y = np.eye(3)[iris.target]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33)
# create and fit es model
from Dawkins import Dawkins
d = Dawkins(n_pop=200, n_generations=2000)
d.fit(X_train, y_train)
d.predict(X_test, y_test)
```