Accelerate Neural Net Training by Progressively Freezing Layers

Overview

FreezeOut

A simple technique to accelerate neural net training by progressively freezing layers.

LRCURVE

This repository contains code for the extended abstract "FreezeOut."

FreezeOut directly accelerates training by annealing layer-wise learning rates to zero on a set schedule, and excluding layers from the backward pass once their learning rate bottoms out.

I had this idea while replying to a reddit comment at 4AM. I threw it in an experiment, and it just worked out of the box (with linear scaling and t_0=0.5), so I went on a 96-hour SCIENCE binge, and now, here we are.

DESIGNCURVE

The exact speedup you get depends on how much error you can tolerate--higher speedups appear to come at the cost of an increase in error, but speedups below 20% should be within a 3% relative error envelope, and speedups around 10% seem to incur no error cost for Scaled Cubic and Unscaled Linear strategies.

Installation

To run this script, you will need PyTorch and a CUDA-capable GPU. If you wish to run it on CPU, just remove all the .cuda() calls.

Running

To run with default parameters, simply call

python train.py

This will by default download CIFAR-100, split it into train, valid, and test sets, then train a k=12 L=76 DenseNet-BC using SGD with Nesterov Momentum.

This script supports command line arguments for a variety of parameters, with the FreezeOut specific parameters being:

  • how_scale selects which annealing strategy to use, among linear, squared, and cubic. Cubic by default.
  • scale_lr determines whether to scale initial learning rates based on t_i. True by default.
  • t_0 is a float between 0 and 1 that decides how far into training to freeze the first layer. 0.8 (pre-cubed) by default.
  • const_time is an experimental setting that increases the number of epochs based on the estimated speedup, in order to match the total training time against a non-FreezeOut baseline. I have not validated if this is worthwhile or not.

You can also set the name of the weights and the metrics log, which model to use, how many epochs to train for, etc.

If you want to calculate an estimated speedup for a given strategy and t_0 value, use the calc_speedup() function in utils.py.

Notes

If you know how to implement this in a static-graph framework (specifically TensorFlow or Caffe2), shoot me an email! It's really easy to do with dynamic graphs, but I believe it to be possible with some simple conditionals in a static graph.

There's (at least) one typo in the paper where it defines the learning rate schedule, there should be a 1/2 in front of alpha.

Acknowledgments

You might also like...
A Neural Net Training Interface on TensorFlow, with focus on speed + flexibility
A Neural Net Training Interface on TensorFlow, with focus on speed + flexibility

Tensorpack is a neural network training interface based on TensorFlow. Features: It's Yet Another TF high-level API, with speed, and flexibility built

A Neural Net Training Interface on TensorFlow, with focus on speed + flexibility
A Neural Net Training Interface on TensorFlow, with focus on speed + flexibility

Tensorpack is a neural network training interface based on TensorFlow. Features: It's Yet Another TF high-level API, with speed, and flexibility built

Simple codebase for flexible neural net training

neural-modular Simple codebase for flexible neural net training. Allows for seamless exchange of models, dataset, and optimizers. Uses hydra for confi

(CVPR2021) ClassSR: A General Framework to Accelerate Super-Resolution Networks by Data Characteristic

ClassSR (CVPR2021) ClassSR: A General Framework to Accelerate Super-Resolution Networks by Data Characteristic Paper Authors: Xiangtao Kong, Hengyuan

GT4SD, an open-source library to accelerate hypothesis generation in the scientific discovery process.
GT4SD, an open-source library to accelerate hypothesis generation in the scientific discovery process.

The GT4SD (Generative Toolkit for Scientific Discovery) is an open-source platform to accelerate hypothesis generation in the scientific discovery process. It provides a library for making state-of-the-art generative AI models easier to use.

a reccurrent neural netowrk that when trained on a peice of text and fed a starting prompt will write its on 250 character text using LSTM layers

RNN-Playwrite a reccurrent neural netowrk that when trained on a peice of text and fed a starting prompt will write its on 250 character text using LS

Neural-net-from-scratch - A simple Neural Network from scratch in Python using the Pymathrix library

A Simple Neural Network from scratch A Simple Neural Network from scratch in Pyt

Meta Language-Specific Layers in Multilingual Language Models

Meta Language-Specific Layers in Multilingual Language Models This repo contains the source codes for our paper On Negative Interference in Multilingu

 Improving Deep Network Debuggability via Sparse Decision Layers
Improving Deep Network Debuggability via Sparse Decision Layers

Improving Deep Network Debuggability via Sparse Decision Layers This repository contains the code for our paper: Leveraging Sparse Linear Layers for D

Comments
  • just a note.

    just a note.

    not a bad idea at all, Dont you think the tensorflow / pytorch community would like to improve it with your idea? These days if you can compress video 3% more then it has a huge impact on data storage centers. If you can train an AI faster, 10%,... boy that's huge... think of the impact on data centers, training time. You should draw some attention from nvidea / intel.

    opened by PGTBoos 0
  • Error in densenet lr update formula:

    Error in densenet lr update formula:

    The 0.05 in the update LR formula overrides the initialized LR per layer (the 1e-1). The line that inits the LR per layer with the scale_lr option is overridden by the formula. m.lr = 1e-1 / m.lr_ratio if self.scale_lr else 1e-1

    The formula with the change: self.optim.param_groups[i]['lr'] = (m.lr)*(1+np.cos(np.pi*self.j/m.max_j))

    opened by nivdu 0
  • fashionMNIST results

    fashionMNIST results

    This code is linked from the fashion-mnist repo, w/ very good results. Do you have a script somewhere I might be able to use to reproduce those numbers?

    Thanks Ben

    opened by bkj 5
  • error

    error

    hi,,,,Im sorry to bother you again....there are still some errors....... like this: file train.py ,line 283, in main train_test(**vars(args))

    filenotfoundError:[Error 2] No such file or directory:Path('logs/densenet_k12L76_ice80_cubicTrue_seed0_epochs100c100_log.jsonl') looking forward to your reply

    opened by jingdonglin 1
Owner
Andy Brock
Dimensionality Diabolist
Andy Brock
2021-MICCAI-Progressively Normalized Self-Attention Network for Video Polyp Segmentation

2021-MICCAI-Progressively Normalized Self-Attention Network for Video Polyp Segmentation Authors: Ge-Peng Ji*, Yu-Cheng Chou*, Deng-Ping Fan, Geng Che

Ge-Peng Ji (Daniel) 85 Dec 30, 2022
Repo for WWW 2022 paper: Progressively Optimized Bi-Granular Document Representation for Scalable Embedding Based Retrieval

BiDR Repo for WWW 2022 paper: Progressively Optimized Bi-Granular Document Representation for Scalable Embedding Based Retrieval. Requirements torch==

Microsoft 11 Oct 20, 2022
The Medical Detection Toolkit contains 2D + 3D implementations of prevalent object detectors such as Mask R-CNN, Retina Net, Retina U-Net, as well as a training and inference framework focused on dealing with medical images.

The Medical Detection Toolkit contains 2D + 3D implementations of prevalent object detectors such as Mask R-CNN, Retina Net, Retina U-Net, as well as a training and inference framework focused on dealing with medical images.

MIC-DKFZ 1.2k Jan 4, 2023
U-2-Net: U Square Net - Modified for paired image training of style transfer

U2-Net: U Square Net Modified for paired image training of style transfer This is an unofficial repo making use of the code which was made available b

Doron Adler 43 Oct 3, 2022
Neural networks applied in recognizing guitar chords using python, AutoML.NET with C# and .NET Core

Chord Recognition Demo application The demo application is written in C# with .NETCore. As of July 9, 2020, the only version available is for windows

Andres Mauricio Rondon Patiño 24 Oct 22, 2022
PyTorch common framework to accelerate network implementation, training and validation

pytorch-framework PyTorch common framework to accelerate network implementation, training and validation. This framework is inspired by works from MML

Dongliang Cao 3 Dec 19, 2022
🏎️ Accelerate training and inference of 🤗 Transformers with easy to use hardware optimization tools

Hugging Face Optimum ?? Optimum is an extension of ?? Transformers, providing a set of performance optimization tools enabling maximum efficiency to t

Hugging Face 842 Dec 30, 2022
Bayesian-Torch is a library of neural network layers and utilities extending the core of PyTorch to enable the user to perform stochastic variational inference in Bayesian deep neural networks

Bayesian-Torch is a library of neural network layers and utilities extending the core of PyTorch to enable the user to perform stochastic variational inference in Bayesian deep neural networks. Bayesian-Torch is designed to be flexible and seamless in extending a deterministic deep neural network architecture to corresponding Bayesian form by simply replacing the deterministic layers with Bayesian layers.

Intel Labs 210 Jan 4, 2023
U^2-Net - Portrait matting This repository explores possibilities of using the original u^2-net model for portrait matting.

U^2-Net - Portrait matting This repository explores possibilities of using the original u^2-net model for portrait matting.

Dennis Bappert 104 Nov 25, 2022
RGBD-Net - This repository contains a pytorch lightning implementation for the 3DV 2021 RGBD-Net paper.

[3DV 2021] We propose a new cascaded architecture for novel view synthesis, called RGBD-Net, which consists of two core components: a hierarchical depth regression network and a depth-aware generator network.

Phong Nguyen Ha 4 May 26, 2022