FasterAI: A library to make smaller and faster models with FastAI.

Last update: Jan 01, 2023

Overview

Fasterai

fasterai is a library created to make neural network smaller and faster. It essentially relies on common compression techniques for networks such as pruning, knowledge distillation, ...

Project Documentation

Visit Read The Docs Project Page or read following README to know more about using Fasterai

Available Methods

1. Pruning

Make your model sparse (i.e. prune it) according to a:

Sparsity: the amount of weights that will be replaced by 0
Granularity: the granularity at which you operate the pruning (removing weights, vectors, kernels, filters)
Method: prune either each layer independantly (local pruning) or the whole model (global pruning)
Criteria: the criteria used to select the weights to remove (magnitude, movement, ...)
Schedule: which schedule you want to use for pruning (one shot, iterative, gradual, ...)

2. Knowledge Distillation

Distill the knowledge acquired by a big model into a smaller one.

3. Lottery Ticket Hypothesis

Find the winning ticket in you network, i.e. the initial subnetwork able to attain at least similar performances than the network as a whole.

Quick Start

1. Create your model with fastai

learn = cnn_learner(dls, model)

2. Get you Fasterai Callback

sp_cb=SparsifyCallback(end_sparsity, granularity, method, criteria, sched_func)

3. Train you model to make it sparse !

learn.fit_one_cycle(n_epochs, cbs=sp_cb)

More about other methods in the tutorials section

Installation

pip install git+https://github.com/nathanhubens/fasterai.git

pip install fasterai

Citing

@misc{Hubens:2020,
  Author = {Nathan Hubens},
  Title = {Fasterai},
  Year = {2020},
  Publisher = {GitHub},
  Journal = {GitHub repository},
  Howpublished = {\url{https://github.com/nathanhubens/fasterai}}
}

You might also like...

shufflev2-yolov5：lighter, faster and easier to deploy

shufflev2-yolov5: lighter, faster and easier to deploy. Evolved from yolov5 and the size of model is only 1.7M (int8) and 3.3M (fp16). It can reach 10+ FPS on the Raspberry Pi 4B when the input size is 320×320~

1.5k Jan 5, 2023

LF-YOLO (Lighter and Faster YOLO) is used to detect defect of X-ray weld image.

This project is based on ultralytics/yolov3. LF-YOLO (Lighter and Faster YOLO) is used to detect defect of X-ray weld image. Download $ git clone http

26 Dec 13, 2022

A Python package for faster, safer, and simpler ML processes

Bender 🤖 A Python package for faster, safer, and simpler ML processes. Why use bender? Bender will make your machine learning processes, faster, safe

6 Dec 13, 2022

Much faster than SORT(Simple Online and Realtime Tracking), a little worse than SORT

QSORT QSORT(Quick + Simple Online and Realtime Tracking) is a simple online and realtime tracking algorithm for 2D multiple object tracking in video s

8 Jul 27, 2022

This project is based on RIFE and aims to make RIFE more practical for users by adding various features and design new models

This project is based on RIFE and aims to make RIFE more practical for users by adding various features and design new models. Because improving the PSNR index is not compatible with subjective effects, we hope this part of work and our academic research are independent of each other.

190 Jan 8, 2023

LeViT a Vision Transformer in ConvNet's Clothing for Faster Inference

FasterAI: A library to make smaller and faster models with FastAI.

Related tags

Overview

Fasterai

Project Documentation

Available Methods

1. Pruning

2. Knowledge Distillation

3. Lottery Ticket Hypothesis

Quick Start

1. Create your model with fastai

2. Get you Fasterai Callback

3. Train you model to make it sparse !

Installation

Citing

You might also like...

shufflev2-yolov5：lighter, faster and easier to deploy

LF-YOLO (Lighter and Faster YOLO) is used to detect defect of X-ray weld image.

A Python package for faster, safer, and simpler ML processes

Much faster than SORT(Simple Online and Realtime Tracking), a little worse than SORT

This project is based on RIFE and aims to make RIFE more practical for users by adding various features and design new models

LeViT a Vision Transformer in ConvNet's Clothing for Faster Inference

Code to run experiments in SLOE: A Faster Method for Statistical Inference in High-Dimensional Logistic Regression.

PyTorch implementation of the Quasi-Recurrent Neural Network - up to 16 times faster than NVIDIA's cuDNN LSTM

A pytorch implementation of faster RCNN detection framework (Use detectron2, it's a masterpiece)

Releases(v0.1.9)

v0.1.9(Jul 4, 2022)

v0.1.8(May 29, 2022)

v0.1.6(Apr 19, 2022)

Owner

Nathan Hubens

SGoLAM - Simultaneous Goal Localization and Mapping

Finding Donors for CharityML

Single-Stage 6D Object Pose Estimation, CVPR 2020

An Ensemble of CNN (Python 3.5.1 Tensorflow 1.3 numpy 1.13)

A simple Tensorflow based library for deep and/or denoising AutoEncoder.

Measures input lag without dedicated hardware, performing motion detection on recorded or live video

Pytorch implementation of Value Iteration Networks (NIPS 2016 best paper)

[CVPR 2022 Oral] Rethinking Minimal Sufficient Representation in Contrastive Learning

Real-ESRGAN aims at developing Practical Algorithms for General Image Restoration.

🔥 Cannlytics-powered artificial intelligence 🤖

The official implementation of CSG-Stump: A Learning Friendly CSG-Like Representation for Interpretable Shape Parsing

You Only 👀 One Sequence

Official implementation of "Intrinsic Dimension, Persistent Homology and Generalization in Neural Networks", NeurIPS 2021.

A curated list of references for MLOps

Paddle implementation for "Highly Efficient Knowledge Graph Embedding Learning with Closed-Form Orthogonal Procrustes Analysis" (NAACL 2021)

Colossal-AI: A Unified Deep Learning System for Large-Scale Parallel Training

Inflated i3d network with inception backbone, weights transfered from tensorflow

Python scripts for performing stereo depth estimation using the MobileStereoNet model in ONNX

:boar: :bear: Deep Learning based Python Library for Stock Market Prediction and Modelling

Implementation of Kaneko et al.'s MaskCycleGAN-VC model for non-parallel voice conversion.