fastai 2.7.14

fastai simplifies training fast and accurate neural nets using modern best practices

08-09-2021 209 words One minute 0 views

Contents

fastai simplifies training fast and accurate neural nets using modern best practices

Stars: 25544, Watchers: 25544, Forks: 7502, Open Issues: 213

The fastai/fastai repo was created 6 years ago and the last code push was 4 days ago.
The project is extremely popular with a mindblowing 25544 github stars!

How to Install fastai

You can install fastai using pip

pip install fastai

or add it to a project with poetry

poetry add fastai

Package Details

Author: Jeremy Howard, Sylvain Gugger, and contributors
License: Apache Software License 2.0
Homepage: https://github.com/fastai/fastai
PyPi:: https://pypi.org/project/fastai/
GitHub Repo:: https://github.com/fastai/fastai

Classifiers

No fastai pypi packages just yet.

Errors

A list of common fastai errors.

Code Examples

Here are some fastai code examples and snippets.

GitHub Issues

The fastai package has 213 open issues on GitHub

CondaHTTPError: HTTP 404 NOT FOUND for url <https://conda.anaconda.org/fastchan/noarch/platformdirs-3.10.0-pyhd8ed1ab_0.conda>
Simple augmentations should not introduce blur!
Resolve CutMix Deprecation Warning
Minimum Requirement of Python 3.8 & PyTorch 1.10
broken link
Docs - Using Colab link doesn't work
FocalLossFlat causing picking error upon model export
Exception Inception
collab_learner and fit_one_cylce crashes kernel on M2 Macs
notebook_launcher doc example not working
broken link on the Collaborative Filtering Page
FastAI load_learner() issue
IMDB example fails

See more issues on GitHub

Related Packages & Articles

deepspeed 0.14.0

DeepSpeed is a Python package developed by Microsoft that provides a deep learning optimization library designed to scale across multiple GPUs and servers. It is capable of training models with billions or even trillions of parameters, achieving excellent system throughput and efficiently scaling to thousands of GPUs.

DeepSpeed is particularly useful for training and inference of large language models, and it falls under the category of Machine Learning Frameworks and Libraries. It is designed to work with PyTorch and offers system innovations such as Zero Redundancy Optimizer (ZeRO), 3D parallelism, and model-parallelism to enable efficient training of large models.