🔪 Elimination based Lightweight Neural Net with Pretrained Weights

Overview

ELimNet

ELimNet: Eliminating Layers in a Neural Network Pretrained with Large Dataset for Downstream Task

  • Removed top layers from pretrained EfficientNetB0 and ResNet18 to construct lightweight CNN model with less than 1M #params.
  • Assessed on Trash Annotations in Context(TACO) Dataset sampled for 6 classes with 20,851 images.
  • Compared performance with lightweight models generated with Optuna's Neural Architecture Search(NAS) constituted with same convolutional blocks.

Quickstart

Installation

# clone the repository
git clone https://github.com/snoop2head/elimnet

# fetch image dataset and unzip
!wget -cq https://aistages-prod-server-public.s3.amazonaws.com/app/Competitions/000081/data/data.zip
!unzip ./data.zip -d ./

Train

# finetune on the dataset with pretrained model
python train.py --model ./model/efficientnet_b0.yaml

# finetune on the dataset with ElimNet
python train.py --model ./model/efficientnet_b0_elim_3.yaml

Inference

# inference with the lastest ran model
python inference.py --model_dir ./exp/latest/

Performance

Performance is compared with (1) original pretrained model and (2) Optuna NAS constructed models with no pretrained weights.

  • Indicates that top convolutional layers eliminated pretrained CNN models outperforms empty Optuna NAS models generated with same convolutional blocks.
  • Suggests that eliminating top convolutional layers creates lightweight model that shows similar(or better) classifcation performance with original pretrained model.
  • Reduces parameters to 7%(or less) of its original parameters while maintaining(or improving) its performance. Saves inference time by 20% or more by eliminating top convolutional layters.

ELimNet vs Pretrained Models (Train)

[100 epochs] # of Parameters # of Layers Train Validation Test F1
Pretrained EfficientNet B0 4.0M 352 Loss: 0.43
Acc: 81.23%
F1: 0.84
Loss: 0.469
Acc: 82.17%
F1: 0.76
0.7493
EfficientNet B0 Elim 2 0.9M 245 Loss:0.652
Acc: 87.22%
F1: 0.84
Loss: 0.622
Acc: 87.22%
F1: 0.77
0.7603
EfficientNet B0 Elim 3 0.30M 181 Loss: 0.602
Acc: 78.17%
F1: 0.74
Loss: 0.661
Acc: 77.41%
F1: 0.74
0.7349
Resnet18 11.17M 69 Loss: 0.578
Acc: 78.90%
F1: 0.76
Loss: 0.700
Acc: 76.17%
F1: 0.719
-
Resnet18 Elim 2 0.68M 37 Loss: 0.447
Acc: 83.73%
F1: 0.71
Loss: 0.712
Acc: 75.42%
F1: 0.71
-

ELimNet vs Pretrained Models (Inference)

# of Parameters # of Layers CPU times (sec) CUDA time (sec) Test Inference Time (sec)
Pretrained EfficientNet B0 4.0M 352 3.9s 4.0s 105.7s
EfficientNet B0 Elim 2 0.9M 245 4.1s 13.0s 83.4s
EfficientNet B0 Elim 3 0.30M 181 3.0s 9.0s 73.5s
Resnet18 11.17M 69 - - -
Resnet18 Elim 2 0.68M 37 - - -

ELimNet vs Empty Optuna NAS Models (Train)

[100 epochs] # of Parameters # of Layers Train Valid Test F1
Empty MobileNet V3 4.2M 227 Loss 0.925
Acc: 65.18%
F1: 0.58
Loss 0.993
Acc: 62.83%
F1: 0.56
-
Empty EfficientNet B0 1.3M 352 Loss 0.867
Acc: 67.28%
F1: 0.61
Loss 0.898
Acc: 66.80%
F1: 0.61
0.6337
Empty DWConv & InvertedResidualv3 NAS 0.08M 66 - Loss: 0.766
Acc: 71.71%
F1: 0.68
0.6740
Empty MBConv NAS 0.33M 141 Loss: 0.786
Acc: 70.72%
F1: 0.66
Loss: 0.866
Acc: 68.09%
F1: 0.62
0.6245
Resnet18 Elim 2 0.68M 37 Loss: 0.447
Acc: 83.73%
F1: 0.71
Loss: 0.712
Acc: 75.42%
F1: 0.71
-
EfficientNet B0 Elim 3 0.30M 181 Loss: 0.602
Acc: 78.17%
F1: 0.74
Loss: 0.661
Acc: 77.41%
F1: 0.74
0.7603

ELimNet vs Empty Optuna NAS Models (Inference)

# of Parameters # of Layers CPU times (sec) CUDA time (sec) Test Inference Time (sec)
Empty MobileNet V3 4.2M 227 4 13 -
Empty EfficientNet B0 1.3M 352 3.780 3.782 68.4s
Empty DWConv &
InvertedResidualv3 NAS
0.08M 66 1 3.5 61.1s
Empty MBConv NAS 0.33M 141 2.14 7.201 67.1s
Resnet18 Elim 2 0.68M 37 - - -
EfficientNet B0 Elim 3 0.30M 181 3.0s 9s 73.5s

Background & WiP

Background

Work in Progress

  • Will test the performance of replacing convolutional blocks with pretrained weights with a single convolutional layer without pretrained weights.
  • Will add ResNet18's inference time data and compare with Optuna's NAS constructed lightweight model.
  • Will test on pretrained MobileNetV3, MnasNet on torchvision with elimination based lightweight model architecture search.
  • Will be applied on other small datasets such as Fashion MNIST dataset and Plant Village dataset.

Others

  • "Empty" stands for model with no pretrained weights.
  • "EfficientNet B0 Elim 2" means 2 convolutional blocks have been eliminated from pretrained EfficientNet B0. Number next to "Elim" annotates how many convolutional blocks have been removed.
  • Table's performance illustrates best performance out of 100 epochs of finetuning on TACO Dataset.

Authors

Owner
snoop2head
break, compose, display
snoop2head
PyTorch implementation of HDN(Homography Decomposition Networks) for planar object tracking

Homography Decomposition Networks for Planar Object Tracking This project is the offical PyTorch implementation of HDN(Homography Decomposition Networ

CaptainHook 48 Dec 15, 2022
structured-generative-modeling

This repository contains the implementation for the paper Information Theoretic StructuredGenerative Modeling, Specially thanks for the open-source co

0 Oct 11, 2021
Python scripts using the Mediapipe models for Halloween.

Mediapipe-Halloween-Examples Python scripts using the Mediapipe models for Halloween. WHY Mainly for fun. But this repository also includes useful exa

Ibai Gorordo 23 Jan 06, 2023
The NEOSSat is a dual-mission microsatellite designed to detect potentially hazardous Earth-orbit-crossing asteroids and track objects that reside in deep space

The NEOSSat is a dual-mission microsatellite designed to detect potentially hazardous Earth-orbit-crossing asteroids and track objects that reside in deep space

John Salib 2 Jan 30, 2022
Syllabus del curso IIC2115 - Programación como Herramienta para la Ingeniería 2022/I

IIC2115 - Programación como Herramienta para la Ingeniería Videos y tutoriales Tutorial CMD Tutorial Instalación Python y Jupyter Tutorial de git-GitH

21 Nov 09, 2022
Code release for "BoxeR: Box-Attention for 2D and 3D Transformers"

BoxeR By Duy-Kien Nguyen, Jihong Ju, Olaf Booij, Martin R. Oswald, Cees Snoek. This repository is an official implementation of the paper BoxeR: Box-A

Nguyen Duy Kien 111 Dec 07, 2022
[CVPR22] Official codebase of Semantic Segmentation by Early Region Proxy.

RegionProxy Figure 2. Performance vs. GFLOPs on ADE20K val split. Semantic Segmentation by Early Region Proxy Yifan Zhang, Bo Pang, Cewu Lu CVPR 2022

Yifan 54 Nov 29, 2022
An exploration of log domain "alternative floating point" for hardware ML/AI accelerators.

This repository contains the SystemVerilog RTL, C++, HLS (Intel FPGA OpenCL to wrap RTL code) and Python needed to reproduce the numerical results in

Facebook Research 373 Dec 31, 2022
[NeurIPS 2021] Source code for the paper "Qu-ANTI-zation: Exploiting Neural Network Quantization for Achieving Adversarial Outcomes"

Qu-ANTI-zation This repository contains the code for reproducing the results of our paper: Qu-ANTI-zation: Exploiting Quantization Artifacts for Achie

Secure AI Systems Lab 8 Mar 26, 2022
Dewarping Document Image By Displacement Flow Estimation with Fully Convolutional Network.

Dewarping Document Image By Displacement Flow Estimation with Fully Convolutional Network

111 Dec 27, 2022
this is a lite easy to use virtual keyboard project for anyone to use

virtual_Keyboard this is a lite easy to use virtual keyboard project for anyone to use motivation I made this for this year's recruitment for RobEn AA

Mohamed Emad 3 Oct 23, 2021
TCNN Temporal convolutional neural network for real-time speech enhancement in the time domain

TCNN Pandey A, Wang D L. TCNN: Temporal convolutional neural network for real-time speech enhancement in the time domain[C]//ICASSP 2019-2019 IEEE Int

凌逆战 16 Dec 30, 2022
7th place solution of Human Protein Atlas - Single Cell Classification on Kaggle

kaggle-hpa-2021-7th-place-solution Code for 7th place solution of Human Protein Atlas - Single Cell Classification on Kaggle. A description of the met

8 Jul 09, 2021
Maximum Spatial Perturbation for Image-to-Image Translation (Official Implementation)

MSPC for I2I This repository is by Yanwu Xu and contains the PyTorch source code to reproduce the experiments in our CVPR2022 paper Maximum Spatial Pe

51 Dec 14, 2022
Official implementation for paper: A Latent Transformer for Disentangled Face Editing in Images and Videos.

A Latent Transformer for Disentangled Face Editing in Images and Videos Official implementation for paper: A Latent Transformer for Disentangled Face

InterDigital 108 Dec 09, 2022
Codebase for Amodal Segmentation through Out-of-Task andOut-of-Distribution Generalization with a Bayesian Model

Codebase for Amodal Segmentation through Out-of-Task andOut-of-Distribution Generalization with a Bayesian Model

Yihong Sun 12 Nov 15, 2022
Awesome Weak-Shot Learning

Awesome Weak-Shot Learning In weak-shot learning, all categories are split into non-overlapped base categories and novel categories, in which base cat

BCMI 162 Dec 30, 2022
Super Resolution for images using deep learning.

Neural Enhance Example #1 — Old Station: view comparison in 24-bit HD, original photo CC-BY-SA @siv-athens. As seen on TV! What if you could increase

Alex J. Champandard 11.7k Dec 29, 2022
Code for a seq2seq architecture with Bahdanau attention designed to map stereotactic EEG data from human brains to spectrograms, using the PyTorch Lightning.

stereoEEG2speech We provide code for a seq2seq architecture with Bahdanau attention designed to map stereotactic EEG data from human brains to spectro

15 Nov 11, 2022
MoCoPnet - Deformable 3D Convolution for Video Super-Resolution

MoCoPnet: Exploring Local Motion and Contrast Priors for Infrared Small Target Super-Resolution Pytorch implementation of local motion and contrast pr

Xinyi Ying 28 Dec 15, 2022