/pytorch-deeplab-xception

DeepLab v3+ model in PyTorch. Support different backbones.

Primary LanguagePythonMIT LicenseMIT

pytorch-deeplab-xception

Update on 2018/12/06. Provide model trained on VOC and SBD datasets.

Update on 2018/11/24. Release newest version code, which fix some previous issues and also add support for new backbones and multi-gpu training. For previous code, please see in previous branch

TODO

  • Support different backbones
  • Support VOC, SBD, Cityscapes and COCO datasets
  • Multi-GPU training
Backbone train/eval os mIoU in val Pretrained Model
ResNet 16/16 78.43% google drive
MobileNet 16/16 70.81% google drive
DRN 16/16 78.87% google drive

Introduction

This is a PyTorch(0.4.1) implementation of DeepLab-V3-Plus. It can use Modified Aligned Xception and ResNet as backbone. Currently, we train DeepLab V3 Plus using Pascal VOC 2012, SBD and Cityscapes datasets.

Results

Installation

The code was tested with Anaconda and Python 3.6. After installing the Anaconda environment:

  1. Clone the repo:

    git clone https://github.com/jfzhang95/pytorch-deeplab-xception.git
cd pytorch-deeplab-xception

  2. Install dependencies:

    For PyTorch dependency, see pytorch.org for more details.

    For custom dependencies:

    pip install matplotlib pillow tensorboardX tqdm

Training

Follow steps below to train your model:

  1. Configure your dataset path in mypath.py.

  2. Input arguments: (see full input arguments via python train.py --help):

    usage: train.py [-h] [--backbone {resnet,xception,drn,mobilenet}]
            [--out-stride OUT_STRIDE] [--dataset {pascal,coco,cityscapes}]
            [--use-sbd] [--workers N] [--base-size BASE_SIZE]
            [--crop-size CROP_SIZE] [--sync-bn SYNC_BN]
            [--freeze-bn FREEZE_BN] [--loss-type {ce,focal}] [--epochs N]
            [--start_epoch N] [--batch-size N] [--test-batch-size N]
            [--use-balanced-weights] [--lr LR]
            [--lr-scheduler {poly,step,cos}] [--momentum M]
            [--weight-decay M] [--nesterov] [--no-cuda]
            [--gpu-ids GPU_IDS] [--seed S] [--resume RESUME]
            [--checkname CHECKNAME] [--ft] [--eval-interval EVAL_INTERVAL]
            [--no-val]

  3. To train deeplabv3+ using Pascal VOC dataset and ResNet as backbone:

    bash train_voc.sh

  4. To train deeplabv3+ using COCO dataset and ResNet as backbone:

    bash train_coco.sh

Acknowledgement

PyTorch-Encoding

Synchronized-BatchNorm-PyTorch

drn