A memory-efficient implementation of DenseNets

Overview

efficient_densenet_pytorch

A PyTorch >=1.0 implementation of DenseNets, optimized to save GPU memory.

Recent updates

  1. Now works on PyTorch 1.0! It uses the checkpointing feature, which makes this code WAY more efficient!!!

Motivation

While DenseNets are fairly easy to implement in deep learning frameworks, most implmementations (such as the original) tend to be memory-hungry. In particular, the number of intermediate feature maps generated by batch normalization and concatenation operations grows quadratically with network depth. It is worth emphasizing that this is not a property inherent to DenseNets, but rather to the implementation.

This implementation uses a new strategy to reduce the memory consumption of DenseNets. We use checkpointing to compute the Batch Norm and concatenation feature maps. These intermediate feature maps are discarded during the forward pass and recomputed for the backward pass. This adds 15-20% of time overhead for training, but reduces feature map consumption from quadratic to linear.

This implementation is inspired by this technical report, which outlines a strategy for efficient DenseNets via memory sharing.

Requirements

  • PyTorch >=1.0.0
  • CUDA

Usage

In your existing project: There is one file in the models folder.

If you care about speed, and memory is not an option, pass the efficient=False argument into the DenseNet constructor. Otherwise, pass in efficient=True.

Options:

  • All options are described in the docstrings of the model files
  • The depth is controlled by block_config option
  • efficient=True uses the memory-efficient version
  • If you want to use the model for ImageNet, set small_inputs=False. For CIFAR or SVHN, set small_inputs=True.

Running the demo:

The only extra package you need to install is python-fire:

pip install fire
  • Single GPU:
CUDA_VISIBLE_DEVICES=0 python demo.py --efficient True --data <path_to_folder_with_cifar10> --save <path_to_save_dir>
  • Multiple GPU:
CUDA_VISIBLE_DEVICES=0,1,2 python demo.py --efficient True --data <path_to_folder_with_cifar10> --save <path_to_save_dir>

Options:

  • --depth (int) - depth of the network (number of convolution layers) (default 40)
  • --growth_rate (int) - number of features added per DenseNet layer (default 12)
  • --n_epochs (int) - number of epochs for training (default 300)
  • --batch_size (int) - size of minibatch (default 256)
  • --seed (int) - manually set the random seed (default None)

Performance

A comparison of the two implementations (each is a DenseNet-BC with 100 layers, batch size 64, tested on a NVIDIA Pascal Titan-X):

Implementation Memory cosumption (GB/GPU) Speed (sec/mini batch)
Naive 2.863 0.165
Efficient 1.605 0.207
Efficient (multi-GPU) 0.985 -

Other efficient implementations

Reference

@article{pleiss2017memory,
  title={Memory-Efficient Implementation of DenseNets},
  author={Pleiss, Geoff and Chen, Danlu and Huang, Gao and Li, Tongcheng and van der Maaten, Laurens and Weinberger, Kilian Q},
  journal={arXiv preprint arXiv:1707.06990},
  year={2017}
}
Owner
Geoff Pleiss
Geoff Pleiss
Template repository to build PyTorch projects from source on any version of PyTorch/CUDA/cuDNN.

The Ultimate PyTorch Source-Build Template Translations: 한국어 TL;DR PyTorch built from source can be x4 faster than a naïve PyTorch install. This repos

Joonhyung Lee/이준형 651 Dec 12, 2022
An algorithm that handles large-scale aerial photo co-registration, based on SURF, RANSAC and PyTorch autograd.

An algorithm that handles large-scale aerial photo co-registration, based on SURF, RANSAC and PyTorch autograd.

Luna Yue Huang 41 Oct 29, 2022
[CVPR 2022] Official Pytorch code for OW-DETR: Open-world Detection Transformer

OW-DETR: Open-world Detection Transformer (CVPR 2022) [Paper] Akshita Gupta*, Sanath Narayan*, K J Joseph, Salman Khan, Fahad Shahbaz Khan, Mubarak Sh

Akshita Gupta 127 Dec 27, 2022
Code for Multinomial Diffusion

Code for Multinomial Diffusion Abstract Generative flows and diffusion models have been predominantly trained on ordinal data, for example natural ima

104 Jan 04, 2023
Official implementation of the RAVE model: a Realtime Audio Variational autoEncoder

RAVE: Realtime Audio Variational autoEncoder Official implementation of RAVE: A variational autoencoder for fast and high-quality neural audio synthes

ACIDS 587 Jan 01, 2023
Assessing the Influence of Models on the Performance of Reinforcement Learning Algorithms applied on Continuous Control Tasks

Assessing the Influence of Models on the Performance of Reinforcement Learning Algorithms applied on Continuous Control Tasks This is the master thesi

Giacomo Arcieri 1 Mar 21, 2022
Experimenting with computer vision techniques to generate annotated image datasets from gameplay recordings automatically.

Experimenting with computer vision techniques to generate annotated image datasets from gameplay recordings automatically. The collected data will then be used to train a deep neural network that can

Martin Valchev 3 Apr 24, 2022
Code for paper "Learning to Reweight Examples for Robust Deep Learning"

learning-to-reweight-examples Code for paper Learning to Reweight Examples for Robust Deep Learning. [arxiv] Environment We tested the code on tensorf

Uber Research 261 Jan 01, 2023
Ratatoskr: Worcester Tech's conference scheduling system

Ratatoskr: Worcester Tech's conference scheduling system In Norse mythology, Ratatoskr is a squirrel who runs up and down the world tree Yggdrasil to

4 Dec 22, 2022
A PyTorch implementation of EfficientDet.

A PyTorch impl of EfficientDet faithful to the original Google impl w/ ported weights

Ross Wightman 1.4k Jan 07, 2023
Implementation of character based convolutional neural network

Character Based CNN This repo contains a PyTorch implementation of a character-level convolutional neural network for text classification. The model a

Ahmed BESBES 248 Nov 21, 2022
Repo for EMNLP 2021 paper "Beyond Preserved Accuracy: Evaluating Loyalty and Robustness of BERT Compression"

beyond-preserved-accuracy Repo for EMNLP 2021 paper "Beyond Preserved Accuracy: Evaluating Loyalty and Robustness of BERT Compression" How to implemen

Kevin Canwen Xu 10 Dec 23, 2022
Code for the paper One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation, CVPR 2021.

One Thing One Click One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation (CVPR2021) Code for the paper One Thi

44 Dec 12, 2022
Dynamic Realtime Animation Control

Our project is targeted at making an application that dynamically detects the user’s expressions and gestures and projects it onto an animation software which then renders a 2D/3D animation realtime

Harsh Avinash 10 Aug 01, 2022
A Distributional Approach To Controlled Text Generation

A Distributional Approach To Controlled Text Generation This is the repository code for the ICLR 2021 paper "A Distributional Approach to Controlled T

NAVER 102 Jan 07, 2023
Aquarius - Enabling Fast, Scalable, Data-Driven Virtual Network Functions

Aquarius Aquarius - Enabling Fast, Scalable, Data-Driven Virtual Network Functions NOTE: We are currently going through the open-source process requir

Zhiyuan YAO 0 Jun 02, 2022
Implementation of H-Transformer-1D, Hierarchical Attention for Sequence Learning using 🤗 transformers

hierarchical-transformer-1d Implementation of H-Transformer-1D, Hierarchical Attention for Sequence Learning using 🤗 transformers In Progress!! 2021.

MyungHoon Jin 7 Nov 06, 2022
This respository includes implementations on Manifoldron: Direct Space Partition via Manifold Discovery

Manifoldron: Direct Space Partition via Manifold Discovery This respository includes implementations on Manifoldron: Direct Space Partition via Manifo

dayang_wang 4 Apr 28, 2022
Supplementary materials to "Spin-optomechanical quantum interface enabled by an ultrasmall mechanical and optical mode volume cavity" by H. Raniwala, S. Krastanov, M. Eichenfield, and D. R. Englund, 2022

Supplementary materials to "Spin-optomechanical quantum interface enabled by an ultrasmall mechanical and optical mode volume cavity" by H. Raniwala,

Stefan Krastanov 1 Jan 17, 2022
GenGNN: A Generic FPGA Framework for Graph Neural Network Acceleration

GenGNN: A Generic FPGA Framework for Graph Neural Network Acceleration Stefan Abi-Karam*, Yuqi He*, Rishov Sarkar*, Lakshmi Sathidevi, Zihang Qiao, Co

Sharc-Lab 19 Dec 15, 2022