Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/jfzhang95/pytorch-deeplab-xception
DeepLab v3+ model in PyTorch. Support different backbones.
https://github.com/jfzhang95/pytorch-deeplab-xception
deeplab-v3-plus drn mobilenetv2 pytorch resnet xception
Last synced: about 1 month ago
JSON representation
DeepLab v3+ model in PyTorch. Support different backbones.
- Host: GitHub
- URL: https://github.com/jfzhang95/pytorch-deeplab-xception
- Owner: jfzhang95
- License: mit
- Created: 2018-06-15T02:07:37.000Z (about 6 years ago)
- Default Branch: master
- Last Pushed: 2022-11-16T04:58:33.000Z (almost 2 years ago)
- Last Synced: 2024-05-13T15:32:44.929Z (4 months ago)
- Topics: deeplab-v3-plus, drn, mobilenetv2, pytorch, resnet, xception
- Language: Python
- Homepage:
- Size: 917 KB
- Stars: 2,863
- Watchers: 46
- Forks: 772
- Open Issues: 129
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- awesome-neural-art - DeepLab V3+ - Implementations of an encoder-decoder Modified Aligned Xception and ResNet as backbone [(paper)](https://arxiv.org/pdf/1802.02611.pdf). (Background subtraction)
README
# 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
- [x] Support different backbones
- [x] Support VOC, SBD, Cityscapes and COCO datasets
- [x] Multi-GPU training
| Backbone | train/eval os |mIoU in val |Pretrained Model|
| :-------- | :------------: |:---------: |:--------------:|
| ResNet | 16/16 | 78.43% | [google drive](https://drive.google.com/open?id=1NwcwlWqA-0HqAPk3dSNNPipGMF0iS0Zu) |
| MobileNet | 16/16 | 70.81% | [google drive](https://drive.google.com/open?id=1G9mWafUAj09P4KvGSRVzIsV_U5OqFLdt) |
| DRN | 16/16 | 78.87% | [google drive](https://drive.google.com/open?id=131gZN_dKEXO79NknIQazPJ-4UmRrZAfI) |
### Introduction
This is a PyTorch(0.4.1) implementation of [DeepLab-V3-Plus](https://arxiv.org/pdf/1802.02611). It
can use Modified Aligned Xception and ResNet as backbone. Currently, we train DeepLab V3 Plus
using Pascal VOC 2012, SBD and Cityscapes datasets.
### Installation
The code was tested with Anaconda and Python 3.6. After installing the Anaconda environment:
0. Clone the repo:
```Shell
git clone https://github.com/jfzhang95/pytorch-deeplab-xception.git
cd pytorch-deeplab-xception
```
1. Install dependencies:
For PyTorch dependency, see [pytorch.org](https://pytorch.org/) for more details.
For custom dependencies:
```Shell
pip install matplotlib pillow tensorboardX tqdm
```
### Training
Follow steps below to train your model:
0. Configure your dataset path in [mypath.py](https://github.com/jfzhang95/pytorch-deeplab-xception/blob/master/mypath.py).
1. Input arguments: (see full input arguments via python train.py --help):
```Shell
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]
```
2. To train deeplabv3+ using Pascal VOC dataset and ResNet as backbone:
```Shell
bash train_voc.sh
```
3. To train deeplabv3+ using COCO dataset and ResNet as backbone:
```Shell
bash train_coco.sh
```
### Acknowledgement
[PyTorch-Encoding](https://github.com/zhanghang1989/PyTorch-Encoding)
[Synchronized-BatchNorm-PyTorch](https://github.com/vacancy/Synchronized-BatchNorm-PyTorch)
[drn](https://github.com/fyu/drn)