🎯 A comprehensive gradient-free optimization framework written in Python

Last update: Dec 26, 2022

Overview

Solid is a Python framework for gradient-free optimization.

It contains basic versions of many of the most common optimization algorithms that do not require the calculation of gradients, and allows for very rapid development using them.

It's a very versatile library that's great for learning, modifying, and of course, using out-of-the-box.

See the detailed documentation here.

Current Features:

Usage:

pip install solidpy
Import the relevant algorithm
Create a class that inherits from that algorithm, and that implements the necessary abstract methods
Call its .run() method, which always returns the best solution and its objective function value

Example:

from random import choice, randint, random
from string import lowercase
from Solid.EvolutionaryAlgorithm import EvolutionaryAlgorithm


class Algorithm(EvolutionaryAlgorithm):
    """
    Tries to get a randomly-generated string to match string "clout"
    """
    def _initial_population(self):
        return list(''.join([choice(lowercase) for _ in range(5)]) for _ in range(50))

    def _fitness(self, member):
        return float(sum(member[i] == "clout"[i] for i in range(5)))

    def _crossover(self, parent1, parent2):
        partition = randint(0, len(self.population[0]) - 1)
        return parent1[0:partition] + parent2[partition:]

    def _mutate(self, member):
        if self.mutation_rate >= random():
            member = list(member)
            member[randint(0,4)] = choice(lowercase)
            member = ''.join(member)
        return member


def test_algorithm():
    algorithm = Algorithm(.5, .7, 500, max_fitness=None)
    best_solution, best_objective_value = algorithm.run()

Testing

To run tests, look in the tests folder.

Use pytest; it should automatically find the test files.

Contributing

Feel free to send a pull request if you want to add any features or if you find a bug.

Check the issues tab for some potential things to do.

Comments

Run flake8 in warning only mode on Python 2 and 3

This will help us find and fix the Python 3 syntax errors (print_function, etc.) A step towards the resolution of https://github.com/100/Solid/issues/6

opened by cclauss 6
Simulated annealing: bug in run method
Description of the bug

The run() method of the SimulatedAnnealing class has a bug when the annealing method does not find a better state than the initial one.

When does it happens

The bug happens when the annealing algorithm fails to find a better state than the initial one. This can happen when the maximum number of steps is low or when the initial guess is already very good.

What is the current behaviour

The tuple returned by the run() method is (None, cost_of_initial_state).

How to fix

Add the line

self.best_state = deepcopy(self.current_state)

between L142 and L143.
opened by nelimee 0
Correction of EA and GA for nondeterministic fitness functions

Correction of an issue that occurs when the fitness function is nondeterministic (shuffled cross-validation for example). In the _select_n method, the total fitness is computed according to the stored fitnesses, but the probs variable is computed according to recalculated fitness values. This slight change makes the method use the stored fitnesses at each time, which solves the problem. This also makes the method run much faster (especially when the fitness function has a high complexity) by removing unnecessary calls to _fitness.

opened by miraaitsaada 0
More Algorithms
Of course, more algorithms are always great.

Some suggestions:

Coordinate descent

Ant colony optimization

Differential evolution

Cuckoo search

Cross-entropy method

enhancement help wanted
opened by 100 0
Numerical Stabilitity

It would be good to find all of the instances where the algorithms may be unstable and handle these cases appropriately (such as overflow). Some cases are handled, but there are probably more.
bug help wanted

opened by 100 0
Better Testing?

Currently, the testing just makes sure that the algorithm runs without error on a toy problem.

It would be nice to do something more akin to unit testing, but I'm not quite sure how to do it in this situation since a lot of the testable functionality is provided by the user.
enhancement help wanted question

opened by 100 0

Releases(0.11)

0.11(Jun 19, 2017)

Source code(tar.gz)
Source code(zip)
0.1(Jun 19, 2017)

Source code(tar.gz)
Source code(zip)

Owner

Devin Soni

GitHub Repository https://100.github.io/Solid/

Predicting Axillary Lymph Node Metastasis in Early Breast Cancer Using Deep Learning on Primary Tumor Biopsy Slides

Predicting Axillary Lymph Node Metastasis in Early Breast Cancer Using Deep Learning on Primary Tumor Biopsy Slides Project | This repo is the officia

[email protected]"> 33 Dec 28, 2022

This repository contains answers of the Shopify Summer 2022 Data Science Intern Challenge.

Data-Science-Intern-Challenge This repository contains answers of the Shopify Summer 2022 Data Science Intern Challenge. Summer 2022 Data Science Inte

1 Jan 11, 2022

Progressive Domain Adaptation for Object Detection

Progressive Domain Adaptation for Object Detection Implementation of our paper Progressive Domain Adaptation for Object Detection, based on pytorch-fa

96 Nov 25, 2022

Official Pytorch implementation for video neural representation (NeRV)

NeRV: Neural Representations for Videos (NeurIPS 2021) Project Page | Paper | UVG Data Hao Chen, Bo He, Hanyu Wang, Yixuan Ren, Ser-Nam Lim, Abhinav S

214 Dec 28, 2022

Kaggle-titanic - A tutorial for Kaggle's Titanic: Machine Learning from Disaster competition. Demonstrates basic data munging, analysis, and visualization techniques. Shows examples of supervised machine learning techniques.

Kaggle-titanic This is a tutorial in an IPython Notebook for the Kaggle competition, Titanic Machine Learning From Disaster. The goal of this reposito

800 Dec 15, 2022

Density-aware Single Image De-raining using a Multi-stream Dense Network (CVPR 2018)

DID-MDN Density-aware Single Image De-raining using a Multi-stream Dense Network He Zhang, Vishal M. Patel [Paper Link] (CVPR'18) We present a novel d

224 Dec 12, 2022

Semantic Segmentation for Aerial Imagery using Convolutional Neural Network

This repo has been deprecated because whole things are re-implemented by using Chainer and I did refactoring for many codes. So please check this newe

27 Sep 23, 2022

Learning to Communicate with Deep Multi-Agent Reinforcement Learning in PyTorch

Learning to Communicate with Deep Multi-Agent Reinforcement Learning This is a PyTorch implementation of the original Lua code release. Overview This

297 Dec 12, 2022

Code to produce syntactic representations that can be used to study syntax processing in the human brain

Can fMRI reveal the representation of syntactic structure in the brain? The code base for our paper on understanding syntactic representations in the

4 Dec 18, 2022

Build and run Docker containers leveraging NVIDIA GPUs

NVIDIA Container Toolkit Introduction The NVIDIA Container Toolkit allows users to build and run GPU accelerated Docker containers. The toolkit includ

15.6k Jan 01, 2023

A library that allows for inference on probabilistic models

Bean Machine Overview Bean Machine is a probabilistic programming language for inference over statistical models written in the Python language using

234 Dec 29, 2022

The 1st Place Solution of the Facebook AI Image Similarity Challenge (ISC21) : Descriptor Track.

ISC21-Descriptor-Track-1st The 1st Place Solution of the Facebook AI Image Similarity Challenge (ISC21) : Descriptor Track. You can check our solution

75 Jan 08, 2023

TalkNet 2: Non-Autoregressive Depth-Wise Separable Convolutional Model for Speech Synthesis with Explicit Pitch and Duration Prediction.

TalkNet 2 [WIP] TalkNet 2: Non-Autoregressive Depth-Wise Separable Convolutional Model for Speech Synthesis with Explicit Pitch and Duration Predictio

69 Dec 17, 2022

🎯 A comprehensive gradient-free optimization framework written in Python

Related tags

Overview

Solid is a Python framework for gradient-free optimization.

It contains basic versions of many of the most common optimization algorithms that do not require the calculation of gradients, and allows for very rapid development using them.

It's a very versatile library that's great for learning, modifying, and of course, using out-of-the-box.

See the detailed documentation here.

Current Features:

Usage:

Example:

Testing

Contributing

Comments

Run flake8 in warning only mode on Python 2 and 3

Simulated annealing: bug in run method

Description of the bug

When does it happens

What is the current behaviour

How to fix

Correction of EA and GA for nondeterministic fitness functions

More Algorithms

Numerical Stabilitity

Better Testing?

Releases(0.11)

0.11(Jun 19, 2017)

0.1(Jun 19, 2017)

Owner

Devin Soni

Predicting Axillary Lymph Node Metastasis in Early Breast Cancer Using Deep Learning on Primary Tumor Biopsy Slides

This repository contains answers of the Shopify Summer 2022 Data Science Intern Challenge.

Progressive Domain Adaptation for Object Detection

Official Pytorch implementation for video neural representation (NeRV)

Kaggle-titanic - A tutorial for Kaggle's Titanic: Machine Learning from Disaster competition. Demonstrates basic data munging, analysis, and visualization techniques. Shows examples of supervised machine learning techniques.

Density-aware Single Image De-raining using a Multi-stream Dense Network (CVPR 2018)

Semantic Segmentation for Aerial Imagery using Convolutional Neural Network

Learning to Communicate with Deep Multi-Agent Reinforcement Learning in PyTorch

Code to produce syntactic representations that can be used to study syntax processing in the human brain

Build and run Docker containers leveraging NVIDIA GPUs

A library that allows for inference on probabilistic models

The 1st Place Solution of the Facebook AI Image Similarity Challenge (ISC21) : Descriptor Track.

TalkNet 2: Non-Autoregressive Depth-Wise Separable Convolutional Model for Speech Synthesis with Explicit Pitch and Duration Prediction.

PyMatting: A Python Library for Alpha Matting

Deep Learning Emotion decoding using EEG data from Autism individuals

Simple PyTorch hierarchical models.

Implementation of Kronecker Attention in Pytorch

Out-of-boundary View Synthesis towards Full-frame Video Stabilization

Official implementation of Self-supervised Image-to-text and Text-to-image Synthesis

Implementation of H-Transformer-1D, Hierarchical Attention for Sequence Learning using 🤗 transformers