PyPI 官网下载 | pytorch-ignite-0.5.0.dev20210731.tar.gz

<div align="center"> <!--  --> <img src="assets/logo/ignite_logo_mixed.svg" width=512> <!-- [](https://travis-ci.com/pytorch/ignite) --> |  [](https://github.com/pytorch/ignite/actions) [](https://app.circleci.com/pipelines/github/pytorch/ignite?branch=master)[](https://app.circleci.com/pipelines/github/pytorch/ignite?branch=master) [](https://codecov.io/gh/pytorch/ignite) [](https://pytorch.org/ignite/index.html) | |:--- |  [](https://anaconda.org/pytorch/ignite) [](https://anaconda.org/pytorch/ignite) [](https://pypi.org/project/pytorch-ignite/) [](https://pepy.tech/project/pytorch-ignite) | |  [](https://anaconda.org/pytorch-nightly/ignite) [](https://pypi.org/project/pytorch-ignite/#history)| |  [](https://hub.docker.com/u/pytorchignite) [](https://optuna.org) [](https://github.com/psf/black) | |  [](https://twitter.com/pytorch_ignite) [](https://www.facebook.com/PyTorch-Ignite-Community-105837321694508) [](https://numfocus.org/sponsored-projects/affiliated-projects) [](https://discord.gg/djZtm3EmKj) | |  [](https://github.com/pytorch/ignite/actions?query=workflow%3A%22PyTorch+version+tests%22)| </div> ## TL;DR Ignite is a high-level library to help with training and evaluating neural networks in PyTorch flexibly and transparently. <div align="center"> <a href="https://colab.research.google.com/github/pytorch/ignite/blob/master/assets/tldr/teaser.ipynb"> <img alt="PyTorch-Ignite teaser" src="assets/tldr/pytorch-ignite-teaser.gif" width=532> </a> _Click on the image to see complete code_ </div> ### Features - [Less code than pure PyTorch](https://raw.githubusercontent.com/pytorch/ignite/master/assets/ignite_vs_bare_pytorch.png) while ensuring maximum control and simplicity - Library approach and no program's control inversion - _Use ignite where and when you need_ - Extensible API for metrics, experiment managers, and other components <!-- ############################################################################################################### --> # Table of Contents - [Table of Contents](#table-of-contents) - [Why Ignite?](#why-ignite) - [Simplified training and validation loop](#simplified-training-and-validation-loop) - [Power of Events & Handlers](#power-of-events--handlers) - [Execute any number of functions whenever you wish](#execute-any-number-of-functions-whenever-you-wish) - [Built-in events filtering](#built-in-events-filtering) - [Stack events to share some actions](#stack-events-to-share-some-actions) - [Custom events to go beyond standard events](#custom-events-to-go-beyond-standard-events) - [Out-of-the-box metrics](#out-of-the-box-metrics) - [Installation](#installation) - [Nightly releases](#nightly-releases) - [Docker Images](#docker-images) - [Using pre-built images](#using-pre-built-images) - [Getting Started](#getting-started) - [Documentation](#documentation) - [Additional Materials](#additional-materials) - [Examples](#examples) - [Tutorials](#tutorials) - [Reproducible Training Examples](#reproducible-training-examples) - [Communication](#communication) - [User feedback](#user-feedback) - [Contributing](#contributing) - [Projects using Ignite](#projects-using-ignite) - [Citing Ignite](#citing-ignite) - [About the team & Disclaimer](#about-the-team--disclaimer) <!-- ############################################################################################################### --> # Why Ignite? Ignite is a **library** that provides three high-level features: - Extremely simple engine and event system - Out-of-the-box metrics to easily evaluate models - Built-in handlers to compose training pipeline, save artifacts and log parameters and metrics ## Simplified training and validation loop No more coding `for/while` loops on epochs and iterations. Users instantiate engines and run them. <details> <summary> Example </summary> ```python from ignite.engine import Engine, Events, create_supervised_evaluator from ignite.metrics import Accuracy # Setup training engine: def train_step(engine, batch): # Users can do whatever they need on a single iteration # Eg. forward/backward pass for any number of models, optimizers, etc # ... trainer = Engine(train_step) # Setup single model evaluation engine evaluator = create_supervised_evaluator(model, metrics={"accuracy": Accuracy()}) def validation(): state = evaluator.run(validation_data_loader) # print computed metrics print(trainer.state.epoch, state.metrics) # Run model's validation at the end of each epoch trainer.add_event_handler(Events.EPOCH_COMPLETED, validation) # Start the training trainer.run(training_data_loader, max_epochs=100) ``` </details> ## Power of Events & Handlers The cool thing with handlers is that they offer unparalleled flexibility (compared to, for example, callbacks). Handlers can be any function: e.g. lambda, simple function, class method, etc. Thus, we do not require to inherit from an interface and override its abstract methods which could unnecessarily bulk up your code and its complexity. ### Execute any number of functions whenever you wish <details> <summary> Examples </summary> ```python trainer.add_event_handler(Events.STARTED, lambda _: print("Start training")) # attach handler with args, kwargs mydata = [1, 2, 3, 4] logger = ... def on_training_ended(data): print(f"Training is ended. mydata={data}") # User can use variables from another scope logger.info("Training is ended") trainer.add_event_handler(Events.COMPLETED, on_training_ended, mydata) # call any number of functions on a single event trainer.add_event_handler(Events.COMPLETED, lambda engine: print(engine.state.times)) @trainer.on(Events.ITERATION_COMPLETED) def log_something(engine): print(engine.state.output) ``` </details> ### Built-in events filtering <details> <summary> Examples </summary> ```python # run the validation every