Official PyTorch Implementation of HELP: Hardware-adaptive Efficient Latency Prediction for NAS via Meta-Learning (NeurIPS 2021 Spotlight)

Overview

[NeurIPS 2021 Spotlight] HELP: Hardware-adaptive Efficient Latency Prediction for NAS via Meta-Learning [Paper]

This is Official PyTorch implementation for HELP: Hardware-adaptive Efficient Latency Prediction for NAS via Meta-Learning.

@inproceedings{lee2021help,
    title     = {HELP: Hardware-Adaptive Efficient Latency Prediction for NAS via Meta-Learning},
    author    = {Lee, Hayeon and Lee, Sewoong and Chong, Song and Hwang, Sung Ju},
    booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
    year      = {2021}
} 

Overview

For deployment, neural architecture search should be hardware-aware, in order to satisfy the device-specific constraints (e.g., memory usage, latency and energy consumption) and enhance the model efficiency. Existing methods on hardware-aware NAS collect a large number of samples (e.g., accuracy and latency) from a target device, either builds a lookup table or a latency estimator. However, such approach is impractical in real-world scenarios as there exist numerous devices with different hardware specifications, and collecting samples from such a large number of devices will require prohibitive computational and monetary cost. To overcome such limitations, we propose Hardware-adaptive Efficient Latency Predictor (HELP), which formulates the device-specific latency estimation problem as a meta-learning problem, such that we can estimate the latency of a model's performance for a given task on an unseen device with a few samples. To this end, we introduce novel hardware embeddings to embed any devices considering them as black-box functions that output latencies, and meta-learn the hardware-adaptive latency predictor in a device-dependent manner, using the hardware embeddings. We validate the proposed HELP for its latency estimation performance on unseen platforms, on which it achieves high estimation performance with as few as 10 measurement samples, outperforming all relevant baselines. We also validate end-to-end NAS frameworks using HELP against ones without it, and show that it largely reduces the total time cost of the base NAS method, in latency-constrained settings.

Prerequisites

  • Python 3.8 (Anaconda)
  • PyTorch 1.8.1
  • CUDA 10.2

Hardware spec used for meta-training the proposed HELP model

  • GPU: A single Nvidia GeForce RTX 2080Ti
  • CPU: Intel(R) Xeon(R) Silver 4114 CPU @ 2.20GHz

Installation

$ conda create --name help python=3.8
$ conda activate help
$ conda install pytorch==1.8.1 torchvision cudatoolkit=10.2 -c pytorch
$ pip install nas-bench-201
$ pip install tqdm
$ conda install scipy
$ conda install pyyaml
$ conda install tensorboard

Contents

1. Experiments on NAS-Bench-201 Search Space

2. Experiments on FBNet Search Space

3. Experiments on OFA Search Space

4. Experiments on HAT Search Space

1. Reproduce Main Results on NAS-Bench-201 Search Space

We provide the code to reproduce the main results on NAS-Bench-201 search space as follows:

  • Computing architecture ranking correlation between latencies estimated by HELP and true measured latencies on unseen devices (Table 3).
  • Latency-constrained NAS Results with MetaD2A + HELP on unseen devices (Table 4).
  • Meta-Training HELP model.

1.1. Data Preparation and Model Checkpoint

We include all required datasets and checkpoints in this github repository.

1.2. [Meta-Test] Architecture ranking correlation

You can compute architecture ranking correlation between latencies estimated by HELP and true measured latencies on unseen devices on NAS-Bench-201 search space (Table 3):

$ python main.py --search_space nasbench201 \
		 --mode 'meta-test' \
		 --num_samples 10 \
		 --num_meta_train_sample 900 \
                 --load_path [Path of Checkpoint File] \
		 --meta_train_devices '1080ti_1,1080ti_32,1080ti_256,silver_4114,silver_4210r,samsung_a50,pixel3,essential_ph_1,samsung_s7' \
		 --meta_valid_devices 'titanx_1,titanx_32,titanx_256,gold_6240' \                 
                 --meta_test_devices 'titan_rtx_256,gold_6226,fpga,pixel2,raspi4,eyeriss' 

You can use checkpoint file provided by this git repository ./data/nasbench201/checkpoint/help_max_corr.pt as follows:

$ python main.py --search_space nasbench201 \
		 --mode 'meta-test' \
		 --num_samples 10 \
		 --num_meta_train_sample 900 \
                 --load_path './data/nasbench201/checkpoint/help_max_corr.pt' \
		 --meta_train_devices '1080ti_1,1080ti_32,1080ti_256,silver_4114,silver_4210r,samsung_a50,pixel3,essential_ph_1,samsung_s7' \
		 --meta_valid_devices 'titanx_1,titanx_32,titanx_256,gold_6240' \                 
                 --meta_test_devices 'titan_rtx_256,gold_6226,fpga,pixel2,raspi4,eyeriss' 

or you can use provided script:

$ bash script/run_meta_test_nasbench201.sh [GPU_NUM]

Architecture Ranking Correlation Results (Table 3)

Method # of Training Samples
From Target Device
Desktop GPU
(Titan RTX Batch 256)
Desktop CPU
(Intel Gold 6226)
Mobile
Pixel2
Raspi4 ASIC FPGA Mean
FLOPS - 0.950 0.826 0.765 0.846 0.437 0.900 0.787
Layer-wise Predictor - 0.667 0.866 - - - - 0.767
BRP-NAS 900 0.814 0.796 0.666 0.847 0.811 0.801 0.789
BRP-NAS
(+extra samples)
3200 0.822 0.805 0.693 0.853 0.830 0.828 0.805
HELP (Ours) 10 0.987 0.989 0.802 0.890 0.940 0.985 0.932

1.3. [Meta-Test] Efficient Latency-constrained NAS combined with MetaD2A

You can reproduce latency-constrained NAS results with MetaD2A + HELP on unseen devices on NAS-Bench-201 search space (Table 4):

$ python main.py --search_space nasbench201 --mode 'nas' \
                 --load_path [Path of Checkpoint File] \
                 --sampled_arch_path 'data/nasbench201/arch_generated_by_metad2a.txt' \
                 --nas_target_device [Device] \ 
                 --latency_constraint [Latency Constraint] 

For example, if you use checkpoint file provided by this git repository, then path of checkpoint file is ./data/nasbench201/checkpoint/help_max_corr.pt, if you set target device as CPU Intel Gold 6226 (gold_6226) with batch size 256 and target latency constraint as 11.0 (ms), command is as follows:

$ python main.py --search_space nasbench201 --mode 'nas' \
                 --load_path './data/nasbench201/checkpoint/help_max_corr.pt' \
                 --sampled_arch_path 'data/nasbench201/arch_generated_by_metad2a.txt' \
                 --nas_target_device gold_6226 \ 
                 --latency_constraint 11.0 

or you can use provided script:

$ bash script/run_nas_metad2a.sh [GPU_NUM]

Efficient Latency-constrained NAS Results (Table 4)

Device # of Training Samples
from Target Device
Latency
Constraint (ms)
Latency
(ms)
Accuracy
(%)
Neural Architecture
Config
GPU Titan RTX
(Batch 256)
titan_rtx_256
10 18.0
21.0
25.0
17.8
18.9
24.2
69.7
71.5
71.8
link
link
link
CPU Intel Gold 6226
gold_6226
10 8.0
11.0
14.0
8.0
10.7
14.3
67.3
70.2
72.1
link
link
link
Mobile Pixel2
pixel2
10 14.0
18.0
22.0
13.0
19.0
25.0
69.7
71.8
73.2
link
link
link
ASIC-Eyeriss
eyeriss
10 5.0
7.0
9.0
3.9
5.1
9.1
71.5
71.8
73.5
link
link
link
FPGA
fpga
10 4.0
5.0
6.0
3.8
4.7
7.4
70.2
71.8
73.5
link
link
link

1.4. Meta-Training HELP model

Note that this process is performed only once for all NAS results.

$ python main.py --search_space nasbench201 \
                 --mode 'meta-train' \
                 --num_samples 10 \
                 --num_meta_train_sample 900 \
                 --meta_train_devices '1080ti_1,1080ti_32,1080ti_256,silver_4114,silver_4210r,samsung_a50,pixel3,essential_ph_1,samsung_s7' \
                 --meta_valid_devices 'titanx_1,titanx_32,titanx_256,gold_6240' \           
                 --meta_test_devices 'titan_rtx_256,gold_6226,fpga,pixel2,raspi4,eyeriss' \
                 --exp_name [EXP_NAME] \
                 --seed 3 # e.g.) 1, 2, 3

or you can use provided script:

$ bash script/run_meta_training_nasbench201.sh [GPU_NUM]

The results (checkpoint file, log file etc) are saved in

./results/nasbench201/[EXP_NAME]

2. Reproduce Main Results on FBNet Search Space

We provide the code to reproduce the main results on FBNet search space as follows:

  • Computing architecture ranking correlation between latencies estimated by HELP and true measured latencies on unseen devices (Table 2).
  • Meta-Training HELP model.

2.1. Data Preparation and Model Checkpoint

We include all required datasets and checkpoints in this github repository.

2.2. [Meta-Test] Architecture ranking correlation

You can compute architecture ranking correlation between latencies estimated by HELP and true measured latencies on unseen devices on FBNet search space (Table 2):

$ python main.py --search_space fbnet \
	--mode 'meta-test' \
	--num_samples 10 \
	--num_episodes 4000 \
	--num_meta_train_sample 4000 \
	--load_path './data/fbnet/checkpoint/help_max_corr.pt' \
	--meta_train_devices '1080ti_1,1080ti_32,1080ti_64,silver_4114,silver_4210r,samsung_a50,pixel3,essential_ph_1,samsung_s7' \
	--meta_valid_devices 'titanx_1,titanx_32,titanx_64,gold_6240' \
	--meta_test_devices 'fpga,raspi4,eyeriss'

or you can use provided script:

$ bash script/run_meta_test_fbnet.sh [GPU_NUM]

Architecture Ranking Correlation Results (Table 2)

Method Raspi4 ASIC FPGA Mean
MAML 0.718 0.763 0.727 0.736
Meta-SGD 0.821 0.822 0.776 0.806
HELP (Ours) 0.887 0.943 0.892 0.910

2.3. Meta-Training HELP model

Note that this process is performed only once for all results.

$ python main.py --search_space fbnet \
	--mode 'meta-train' \
	--num_samples 10 \
	--num_episodes 4000 \
	--num_meta_train_sample 4000 \
	--exp_name [EXP_NAME] \
	--meta_train_devices '1080ti_1,1080ti_32,1080ti_64,silver_4114,silver_4210r,samsung_a50,pixel3,essential_ph_1,samsung_s7' \
	--meta_valid_devices 'titanx_1,titanx_32,titanx_64,gold_6240' \
	--meta_test_devices 'fpga,raspi4,eyeriss' \
	--seed 3 # e.g.) 1, 2, 3

or you can use provided script:

$ bash script/run_meta_training_fbnet.sh [GPU_NUM]

The results (checkpoint file, log file etc) are saved in

./results/fbnet/[EXP_NAME]

3. Reproduce Main Results on OFA Search Space

We provide the code to reproduce the main results on OFA search space as follows:

  • Latency-constrained NAS Results with accuracy predictor of OFA + HELP on unseen devices (Table 5).
  • Validating obatined neural architecture on ImageNet-1K.
  • Meta-Training HELP model.

3.1. Data Preparation and Model Checkpoint

We include required datasets except ImageNet-1K, and checkpoints in this github repository. To validate obatined neural architecture on ImageNet-1K, you should download ImageNet-1K (2012 ver.)

3.2. [Meta-Test] Efficient Latency-constrained NAS combined with accuracy predictor of OFA

You can reproduce latency-constrained NAS results with OFA + HELP on unseen devices on OFA search space (Table 5):

python main.py \
	--search_space ofa \
	--mode nas \
	--num_samples 10 \
	--seed 3 \
	--num_meta_train_sample 4000 \
	--load_path './data/ofa/checkpoint/help_max_corr.pt' \
	--nas_target_device [DEVICE_NAME] \
	--latency_constraint [LATENCY_CONSTRAINT] \
	--exp_name 'nas' \
	--meta_train_devices '2080ti_1,2080ti_32,2080ti_64,titan_xp_1,titan_xp_32,titan_xp_64,v100_1,v100_32,v100_64' \
	--meta_valid_devices 'titan_rtx_1,titan_rtx_32' \
	--meta_test_devices 'titan_rtx_64' 

For example,

$ python main.py \
	--search_space ofa \
	--mode nas \
	--num_samples 10 \
	--seed 3 \
	--num_meta_train_sample 4000 \
	--load_path './data/ofa/checkpoint/help_max_corr.pt' \
	--nas_target_device titan_rtx_64 \
	--latency_constraint 20 \
	--exp_name 'nas' \
	--meta_train_devices '2080ti_1,2080ti_32,2080ti_64,titan_xp_1,titan_xp_32,titan_xp_64,v100_1,v100_32,v100_64' \
	--meta_valid_devices 'titan_rtx_1,titan_rtx_32' \
	--meta_test_devices 'titan_rtx_64' 

or you can use provided script:

$ bash script/run_nas_ofa.sh [GPU_NUM]

Efficient Latency-constrained NAS Results (Table 5)

Device Sample from
Target Device
Latency
Constraint (ms)
Latency
(ms)
Accuracy
(%)
Architecture
config
GPU Titan RTX
(Batch 64)
10 20
23
28
20.3
23.1
28.6
76.0
76.8
77.9
link
link
link
CPU Intel Gold 6226 20 170
190
147
171
77.6
78.1
link
link
Jetson AGX Xavier 10 65
70
67.4
76.4
75.9
76.4
link
link

3.3. Validating obtained neural architecture on ImageNet-1K

$ python validate_imagenet.py \
		--config_path [Path of neural architecture config file]
		--imagenet_save_path [Path of ImageNet 1k]

for example,

$ python validate_imagenet.py \
		--config_path 'data/ofa/architecture_config/gpu_titan_rtx_64/latency_28.6ms_accuracy_77.9.json' \
		--imagenet_save_path './ILSVRC2012'

3.4. Meta-training HELP model

Note that this process is performed only once for all results.

$ python main.py --search_space ofa \
		--mode 'meta-train' \
		--num_samples 10 \
		--num_meta_train_sample 4000 \
		--exp_name [EXP_NAME] \
                --meta_train_devices '2080ti_1,2080ti_32,2080ti_64,titan_xp_1,titan_xp_32,titan_xp_64,v100_1,v100_32,v100_64' \
                --meta_valid_devices 'titan_rtx_1,titan_rtx_32' \
                --meta_test_devices 'titan_rtx_64' \
		--seed 3 # e.g.) 1, 2, 3

or you can use provided script:

$ bash script/run_meta_training_ofa.sh [GPU_NUM]

4. Main Results on HAT Search Space

We provide the neural architecture configurations to reproduce the results of machine translation (WMT'14 En-De Task) on HAT search space.

Efficient Latency-constrained NAS Results

Task Device Samples from
Target Device
Latency BLEU score Architecture
Config
WMT'14 En-De GPU NVIDIA Titan RTX 10 74.0ms
106.5ms
27.19
27.44
link
link
WMT'14 En-De CPU Intel Xeon Gold 6240 10 159.6ms
343.2ms
27.20
27.52
link
link

You can test models by BLEU score and Computing Latency.

Reference

Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks (ICML17)

Meta-SGD: Learning to Learn Quickly for Few-Shot Learning

Once-for-All: Train One Network and Specialize it for Efficient Deployment (ICLR20)

NAS-Bench-201: Extending the Scope of Reproducible Neural Architecture Search (ICLR20)

BRP-NAS: Prediction-based NAS using GCNs (NeurIPS20)

HAT: Hardware Aware Transformers for Efficient Natural Language Processing (ACL20)

Rapid Neural Architecture Search by Learning to Generate Graphs from Datasets (ICLR21)

HW-NAS-Bench: Hardware-Aware Neural Architecture Search Benchmark (ICLR21)

Owner
Ph.D. student @ School of Computing, Korea Advanced Institute of Science and Technology (KAIST)
SatelliteNeRF - PyTorch-based Neural Radiance Fields adapted to satellite domain

SatelliteNeRF PyTorch-based Neural Radiance Fields adapted to satellite domain.

Kai Zhang 46 Nov 20, 2022
Feed forward VQGAN-CLIP model, where the goal is to eliminate the need for optimizing the latent space of VQGAN for each input prompt

Feed forward VQGAN-CLIP model, where the goal is to eliminate the need for optimizing the latent space of VQGAN for each input prompt. This is done by

Mehdi Cherti 135 Dec 30, 2022
PyTorch Implementation of Spatially Consistent Representation Learning(SCRL)

Spatially Consistent Representation Learning (CVPR'21) Official PyTorch implementation of Spatially Consistent Representation Learning (SCRL). This re

Kakao Brain 102 Nov 03, 2022
Oriented Response Networks, in CVPR 2017

Oriented Response Networks [Home] [Project] [Paper] [Supp] [Poster] Torch Implementation The torch branch contains: the official torch implementation

ZhouYanzhao 217 Dec 12, 2022
A Kitti Road Segmentation model implemented in tensorflow.

KittiSeg KittiSeg performs segmentation of roads by utilizing an FCN based model. The model achieved first place on the Kitti Road Detection Benchmark

Marvin Teichmann 890 Jan 04, 2023
A project which aims to protect your privacy using inexpensive hardware and easily modifiable software

Protecting your privacy using an ESP32, an IR sensor and a python script This project, which I personally call the "never-gonna-catch-me-in-the-act-ev

8 Oct 10, 2022
SMCA replication There are no extra compiled components in SMCA DETR and package dependencies are minimal

Usage There are no extra compiled components in SMCA DETR and package dependencies are minimal, so the code is very simple to use. We provide instruct

22 May 06, 2022
Multi-query Video Retreival

Multi-query Video Retreival

Princeton Visual AI Lab 17 Nov 22, 2022
This is an official implementation of "Polarized Self-Attention: Towards High-quality Pixel-wise Regression"

Polarized Self-Attention: Towards High-quality Pixel-wise Regression This is an official implementation of: Huajun Liu, Fuqiang Liu, Xinyi Fan and Don

DeLightCMU 212 Jan 08, 2023
Deep Learning pipeline for motor-imagery classification.

BCI-ToolBox 1. Introduction BCI-ToolBox is deep learning pipeline for motor-imagery classification. This repo contains five models: ShallowConvNet, De

DongHee 18 Oct 31, 2022
Learning Visual Words for Weakly-Supervised Semantic Segmentation

[IJCAI 2021] Learning Visual Words for Weakly-Supervised Semantic Segmentation Implementation of IJCAI 2021 paper Learning Visual Words for Weakly-Sup

Lixiang Ru 24 Oct 05, 2022
Graph Analysis From Scratch

Graph Analysis From Scratch Goal In this notebook we wanted to implement some functionalities to analyze a weighted graph only by using algorithms imp

Arturo Ghinassi 0 Sep 17, 2022
An Api for Emotion recognition.

PLAYEMO Playemo was built from the ground-up with Flask, a python tool that makes it easy for developers to build APIs. Use Cases Is Python your langu

greek geek 2 Jul 16, 2022
Simple Baselines for Human Pose Estimation and Tracking

Simple Baselines for Human Pose Estimation and Tracking News Our new work High-Resolution Representations for Labeling Pixels and Regions is available

Microsoft 2.7k Jan 05, 2023
TensorFlow implementation of the algorithm in the paper "Decoupled Low-light Image Enhancement"

Decoupled Low-light Image Enhancement Shijie Hao1,2*, Xu Han1,2, Yanrong Guo1,2 & Meng Wang1,2 1Key Laboratory of Knowledge Engineering with Big Data

17 Apr 25, 2022
PyTorch version repo for CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes

Study-CSRNet-pytorch This is the PyTorch version repo for CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes

0 Mar 01, 2022
Image Segmentation and Object Detection in Pytorch

Image Segmentation and Object Detection in Pytorch Pytorch-Segmentation-Detection is a library for image segmentation and object detection with report

Daniil Pakhomov 732 Dec 10, 2022
Re-implementation of 'Grokking: Generalization beyond overfitting on small algorithmic datasets'

Re-implementation of the paper 'Grokking: Generalization beyond overfitting on small algorithmic datasets' Paper Original paper can be found here Data

Tom Lieberum 38 Aug 09, 2022
Implementation of Ag-Grid component for Streamlit

streamlit-aggrid AgGrid is an awsome grid for web frontend. More information in https://www.ag-grid.com/. Consider purchasing a license from Ag-Grid i

Pablo Fonseca 556 Dec 31, 2022
Real-Time-Student-Attendence-System - Real Time Student Attendence System

Real-Time-Student-Attendence-System The Student Attendance Management System Pro

Rounak Das 1 Feb 15, 2022