Month: September 2021

Google

Gcp-doctor: a command-line diagnostics tool for GCP customers

gcp-doctor is a command-line diagnostics tool for GCP customers. It finds and helps to fix common issues in Google Cloud Platform projects. It is used to test projects against a wide range of best-practices and frequent mistakes, based on the troubleshooting experience of the Google Cloud Support team. gcp-doctor is open-source and contributions are welcome! Note that this is not an officially supported Google product, but a community effort. The Google Cloud Support team maintains this code and we do […]

Read more
Algorithms

Sequential Model-based Algorithm Configuration

Copyright (C) 2016-2018 AutoML Group Attention: This package is a reimplementation of the original SMAC tool (see reference below). However, the reimplementation slightly differs from the original SMAC. For comparisons against the original SMAC, we refer to a stable release of SMAC (v2) in Java which can be found here. The documentation can be found here. Status for master branch: Status for the development branch SMAC is a tool for algorithm configuration to optimize the parameters of arbitrary algorithms across […]

Read more
Generator

Genetic feature selection module for scikit-learn

Genetic feature selection module for scikit-learn Genetic algorithms mimic the process of natural selection to search for optimal values of a function. Installation The easiest way to install sklearn-genetic is using pip pip install sklearn-genetic or conda conda install -c conda-forge sklearn-genetic Requirements Python >= 2.7 scikit-learn >= 0.20.3 DEAP >= 1.0.2 Example from __future__ import print_function import numpy as np from sklearn import datasets, linear_model from genetic_selection import GeneticSelectionCV def main(): iris = datasets.load_iris() # Some    

Read more
Generator

A strongly-typed genetic programming framework for Python

“If an army of monkeys were strumming on typewriters they might write all the books in the British Museum.” monkeys is a framework designed to make genetic programming in Python accessible, quick, flexible, and fun. Get started here. What is genetic programming? Genetic programming algorithms are a class of evolutionary algorithms in which solutions to a problem are represented as executable tree structures – programs. In order to use genetic programming in order to solve a problem, two things must […]

Read more
Data Scientists

Supervised domain-agnostic prediction framework for probabilistic modelling

A supervised domain-agnostic framework that allows for probabilistic modelling, namely the prediction of probability distributions for individual data points. The package offers a variety of features and specifically allows for the implementation of probabilistic prediction strategies in the supervised contexts comparison of frequentist and Bayesian prediction methods strategy optimization through hyperparamter tuning and ensemble methods (e.g. bagging) workflow automation List of developers and contributors Documentation The full documentation is available here. Installation Installation is easy using Python’s package manager $ […]

Read more
Python

Build a Personal Diary With Django and Python

A diary is a personal safe space. With the help of Django, you can create a diary on your own computer without storing data in anyone else’s cloud. By following along with the project below, you’ll see how quickly you can build a functioning web app in Django without any external dependencies. In this tutorial, you’ll learn how to: Set up a Django project Work with the standard SQLite database Make use of the Django admin site Create models and […]

Read more
Facebook NLP Research
NLP

slimIPL: Language-Model-Free Iterative Pseudo-Labeling

Abstract Recent results in end-to-end automatic speech recognition have demonstrated the efficacy of pseudo-labeling for semi-supervised models trained both with Connectionist Temporal Classification (CTC) and Sequence-to-Sequence (seq2seq) losses. Iterative Pseudo-Labeling (IPL), which continuously trains a single model using pseudo-labels iteratively re-generated as the model learns, has been shown to further improve performance in ASR. We improve upon the IPL algorithm: as the model learns, we propose to iteratively re-generate transcriptions with hard labels (the most probable tokens), that is, without […]

Read more
1 … 24 25 26 27 28 … 48