mindcv_fixed/configs/pvt

• Pyramid Vision Transformer
• • Introduction
  • Results
  • • • Notes
      • Notes
  • Quick Start
  • • Preparation
    • • Installation
      • Dataset Preparation
    • Training
    • Validation
    • Deployment (optional)

Pyramid Vision Transformer

Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions

Introduction

---

PVT is a general backbone network for dense prediction without convolution operation. PVT introduces a pyramid structure in Transformer to generate multi-scale feature maps for dense prediction tasks. PVT uses a gradual reduction strategy to control the size of the feature maps through the patch embedding layer, and proposes a spatial reduction attention (SRA) layer to replace the traditional multi head attention layer in the encoder, which greatly reduces the computing/memory overhead.

Results

---

Model	Context	Top-1 (%)	Top-5 (%)	Params (M)	Train T.	Infer T.	Download	Config	Log
PVT_tiny	D910x8-G	74.92			433s/epoch	16ms/step	model	cfg	log
PVT_small	D910x8-G	79.66			538s/epoch	30ms/step	model	cfg	log
PVT_medium	D910x8-G	81.82			766s/epoch	47ms/step	model	cfg	log
PVT_large	D910x8-G	81.75			1074s/epoch	67ms/step	model	cfg	log

Notes

• All models are trained on ImageNet-1K training set and the top-1 accuracy is reported on the validatoin set.
• Context: GPU_TYPE x pieces - G/F, G - graph mode, F - pynative mode with ms function.

Notes

• All models are trained on ImageNet-1K training set and the top-1 accuracy is reported on the validatoin set.
• All models are trained with Ascend910*8 in graph mode and infered with an Ascend310.

Quick Start

---

Preparation

Installation

Please refer to the installation instruction in MindCV.

Dataset Preparation

Please download the ImageNet-1K dataset for model training and validation.

Training

• Hyper-parameters. The hyper-parameter configurations for producing the reported results are stored in the yaml files in mindcv/configs/pvt folder. For example, to train with one of these configurations, you can run:

  # train densenet121 on 8 GPUs
  export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
  mpirun -n 8 python train.py -c configs/pvt/pvt_tiny_ascend.yaml --data_dir /path/to/imagenet
  

  Note that the number of GPUs/Ascends and batch size will influence the training results. To reproduce the training result at most, it is recommended to use the same number of GPUs/Ascneds with the same batch size.

• Finetuning. Here is an example for finetuning a pretrained pvt tiny on CIFAR10 dataset using Momentum optimizer.

  python train.py --model=pvt_tiny --pretrained --opt=momentum --lr=0.001 dataset=cifar10 --num_classes=10 --dataset_download
  

Detailed adjustable parameters and their default value can be seen in config.py.

Validation

• To validate the trained model, you can use validate.py. Here is an example for pvt tiny to verify the accuracy of
  pretrained weights.

  python validate.py --model=model=pvt_tiny  --dataset=imagenet --val_split=val --pretrained
  

• To validate the model, you can use validate.py. Here is an example for pvt tiny to verify the accuracy of your training.

  python validate.py --model=model=pvt_tiny  --dataset=imagenet --val_split=val --ckpt_path='./ckpt/model=pvt_tiny-best.ckpt'
  

Deployment (optional)

Please refer to the deployment tutorial in MindCV.