joeddav/devol
Genetic neural architecture search with Keras
This tool helps machine learning engineers and researchers automatically design the optimal neural network architecture for classification problems. You provide your dataset and define the constraints for your model's structure, and it uses genetic algorithms to evolve and test many different configurations. The output is a highly tuned Keras model that performs well on your specific classification task, along with insights into which architectural choices are most effective.
952 stars. No commits in the last 6 months.
Use this if you need to find the best deep learning model architecture for a classification problem without manually trying countless configurations, and you have significant computational resources available.
Not ideal if you have very limited computational power or are working on problems other than classification, as the process of evaluating many models can be very resource-intensive.
Stars
952
Forks
114
Language
Python
License
MIT
Category
Last pushed
May 25, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/joeddav/devol"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
optuna/optuna
A hyperparameter optimization framework
keras-team/keras-tuner
A Hyperparameter Tuning Library for Keras
KernelTuner/kernel_tuner
Kernel Tuner
syne-tune/syne-tune
Large scale and asynchronous Hyperparameter and Architecture Optimization at your fingertips.
deephyper/deephyper
DeepHyper: A Python Package for Massively Parallel Hyperparameter Optimization in Machine Learning