EfficientNetV2-with-TPU - Cifar-10 case study

Overview

EfficientNetV2-with-TPU

EfficientNet

EfficientNetV2 adalah jenis jaringan saraf convolutional yang memiliki kecepatan pelatihan lebih cepat dan efisiensi parameter yang lebih baik dari model sebelumnya . Untuk mengembangkan model ini, penulis menggunakan kombinasi pencarian dan penskalaan arsitektur saraf yang sadar pelatihan , untuk bersama-sama mengoptimalkan kecepatan pelatihan. Model dicari dari ruang pencarian yang diperkaya dengan operasi baru seperti Fused-MBConv .

Secara arsitektur perbedaan utama adalah:

  • EfficientNetV2 secara ekstensif menggunakan MBConv dan fusi-MBConv yang baru ditambahkan di lapisan awal.
  • EfficientNetV2 lebih memilih rasio ekspansi yang lebih kecil untuk MBConv karena rasio ekspansi yang lebih kecil cenderung memiliki lebih sedikit overhead akses memori.
  • EfficientNetV2 lebih menyukai ukuran kernel 3x3 yang lebih kecil, tetapi menambahkan lebih banyak lapisan untuk mengkompensasi bidang reseptif yang berkurang yang dihasilkan dari ukuran kernel yang lebih kecil.
  • EfficientNetV2 sepenuhnya menghapus tahap stride-1 terakhir di EfficientNet asli, mungkin karena ukuran parameternya yang besar dan overhead akses memori

Note

Model Size acc-val top-5 acc-test weight
EfficientNetV2B0 224 90.68 99.76 89.86 imagenet
EfficientNetV2B1 240 90.76 99.78 90.07 imagenet
EfficientNetV2B2 260 87.08 99.48 86.85 imagenet
EfficientNetV2B3 300 90.38 99.80 89.29 imagenet
EfficientNetV2T 320 92.80 99.86 92.53 imagenet
EfficientNetV2S 384 89.94 99.74 89.27 imagenet
EfficientNetV2M 480 91.86 99.70 90.53 imagenet
EfficientNetV2L 480 93.10 99.80 92.38 imagenet
EfficientNetV2XL 512 93.24 99.72 93.41 imagenet21K-ft1k
  • Train 90%(45000rb)

  • Validation 10%(5000rb)

  • Test(10000rb)

  • Epochs = 25

  • WeightDecay = 1e-5

  • Batchsize = 16 * 8(strategy.num_replicas_in_sync)

  • optimizers adabelief dengan LearningRateSchduler(Triangular2CyclicalLearningRate) dan Rectified = True(mencegah overshoot)

  • cifar-10 tidak di sarankan untuk di ubah ukuran nya, saya mengubah ukuran nya hanya untuk milihat apakah bagus/tidak efficientnetv2 saat mempelajari cifar-10

Referensi

Owner
Sultan syach
Sultan syach
official Pytorch implementation of ICCV 2021 paper FuseFormer: Fusing Fine-Grained Information in Transformers for Video Inpainting.

FuseFormer: Fusing Fine-Grained Information in Transformers for Video Inpainting By Rui Liu, Hanming Deng, Yangyi Huang, Xiaoyu Shi, Lewei Lu, Wenxiu

77 Dec 27, 2022
A Parameter-free Deep Embedded Clustering Method for Single-cell RNA-seq Data

A Parameter-free Deep Embedded Clustering Method for Single-cell RNA-seq Data Overview Clustering analysis is widely utilized in single-cell RNA-seque

AI-Biomed @NSCC-gz 3 May 08, 2022
Official repository for the paper "Self-Supervised Models are Continual Learners" (CVPR 2022)

Self-Supervised Models are Continual Learners This is the official repository for the paper: Self-Supervised Models are Continual Learners Enrico Fini

Enrico Fini 73 Dec 18, 2022
Code for our EMNLP 2021 paper “Heterogeneous Graph Neural Networks for Keyphrase Generation”

GATER This repository contains the code for our EMNLP 2021 paper “Heterogeneous Graph Neural Networks for Keyphrase Generation”. Our implementation is

Jiacheng Ye 12 Nov 24, 2022
ICCV2021 Oral SA-ConvONet: Sign-Agnostic Optimization of Convolutional Occupancy Networks

Sign-Agnostic Convolutional Occupancy Networks Paper | Supplementary | Video | Teaser Video | Project Page This repository contains the implementation

64 Jan 05, 2023
Pytorch implementation of Supporting Clustering with Contrastive Learning, NAACL 2021

Supporting Clustering with Contrastive Learning SCCL (NAACL 2021) Dejiao Zhang, Feng Nan, Xiaokai Wei, Shangwen Li, Henghui Zhu, Kathleen McKeown, Ram

231 Jan 05, 2023
Doing the asl sign language classification on static images using graph neural networks.

SignLangGNN When GNNs 💜 MediaPipe. This is a starter project where I tried to implement some traditional image classification problem i.e. the ASL si

10 Nov 09, 2022
Object Detection using YOLO from PyImageSearch

Object Detection using YOLO from PyImageSearch By applying object detection, you’ll not only be able to determine what is in an image, but also where

Mohamed NIANG 1 Feb 09, 2022
alfred-py: A deep learning utility library for **human**

Alfred Alfred is command line tool for deep-learning usage. if you want split an video into image frames or combine frames into a single video, then a

JinTian 800 Jan 03, 2023
Lbl2Vec learns jointly embedded label, document and word vectors to retrieve documents with predefined topics from an unlabeled document corpus.

Lbl2Vec Lbl2Vec is an algorithm for unsupervised document classification and unsupervised document retrieval. It automatically generates jointly embed

sebis - TUM - Germany 61 Dec 20, 2022
RoboDesk A Multi-Task Reinforcement Learning Benchmark

RoboDesk A Multi-Task Reinforcement Learning Benchmark If you find this open source release useful, please reference in your paper: @misc{kannan2021ro

Google Research 66 Oct 07, 2022
PyTorch implementation for the ICLR 2020 paper "Understanding the Limitations of Variational Mutual Information Estimators"

Smoothed Mutual Information ``Lower Bound'' Estimator PyTorch implementation for the ICLR 2020 paper Understanding the Limitations of Variational Mutu

50 Nov 09, 2022
A system for quickly generating training data with weak supervision

Programmatically Build and Manage Training Data Announcement The Snorkel team is now focusing their efforts on Snorkel Flow, an end-to-end AI applicat

Snorkel Team 5.4k Jan 02, 2023
LabelImg is a graphical image annotation tool.

LabelImgPlus LabelImg is a graphical image annotation tool. This project is not updated with new functions now. More functions are supported with Labe

lzx1413 200 Dec 20, 2022
In this project I played with mlflow, streamlit and fastapi to create a training and prediction app on digits

Fastapi + MLflow + streamlit Setup env. I hope I covered all. pip install -r requirements.txt Start app Go in the root dir and run these Streamlit str

76 Nov 23, 2022
Deep Residual Networks with 1K Layers

Deep Residual Networks with 1K Layers By Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun. Microsoft Research Asia (MSRA). Table of Contents Introduc

Kaiming He 856 Jan 06, 2023
[CVPR 2022 Oral] Crafting Better Contrastive Views for Siamese Representation Learning

Crafting Better Contrastive Views for Siamese Representation Learning (CVPR 2022 Oral) 2022-03-29: The paper was selected as a CVPR 2022 Oral paper! 2

249 Dec 28, 2022
Official code for "Eigenlanes: Data-Driven Lane Descriptors for Structurally Diverse Lanes", CVPR2022

[CVPR 2022] Eigenlanes: Data-Driven Lane Descriptors for Structurally Diverse Lanes Dongkwon Jin, Wonhui Park, Seong-Gyun Jeong, Heeyeon Kwon, and Cha

Dongkwon Jin 106 Dec 29, 2022
Learned model to estimate number of distinct values (NDV) of a population using a small sample.

Learned NDV estimator Learned model to estimate number of distinct values (NDV) of a population using a small sample. The model approximates the maxim

2 Nov 21, 2022
Rate-limit-semaphore - Semaphore implementation with rate limit restriction for async-style (any core)

Rate Limit Semaphore Rate limit semaphore for async-style (any core) There are t

Yan Kurbatov 4 Jun 21, 2022