pytorch-deeplab-xception-master
所属分类:其他
开发工具:Python
文件大小:561KB
下载次数:3
上传日期:2020-08-12 00:30:27
上 传 者:
12334456
说明: pytroch-deeplab实现语义分割,backbone resnet\xception\mobilenet
(pytroch-deeplab semantatic segmentation backbone resnet\xception\mobilenet)
文件列表:
pytorch-deeplab-xception-master (0, 2020-08-11)
pytorch-deeplab-xception-master\dataloaders (0, 2020-08-11)
pytorch-deeplab-xception-master\dataloaders\__init__.py (2008, 2020-08-11)
pytorch-deeplab-xception-master\dataloaders\custom_transforms.py (5050, 2020-08-11)
pytorch-deeplab-xception-master\dataloaders\datasets (0, 2020-08-11)
pytorch-deeplab-xception-master\dataloaders\datasets\__init__.py (0, 2020-08-11)
pytorch-deeplab-xception-master\dataloaders\datasets\cityscapes.py (5370, 2020-08-11)
pytorch-deeplab-xception-master\dataloaders\datasets\coco.py (5718, 2020-08-11)
pytorch-deeplab-xception-master\dataloaders\datasets\combine_dbs.py (3310, 2020-08-11)
pytorch-deeplab-xception-master\dataloaders\datasets\pascal.py (4492, 2020-08-11)
pytorch-deeplab-xception-master\dataloaders\datasets\sbd.py (4081, 2020-08-11)
pytorch-deeplab-xception-master\dataloaders\utils.py (3345, 2020-08-11)
pytorch-deeplab-xception-master\doc (0, 2020-08-11)
pytorch-deeplab-xception-master\doc\deeplab_resnet.py (11247, 2020-08-11)
pytorch-deeplab-xception-master\doc\deeplab_xception.py (16199, 2020-08-11)
pytorch-deeplab-xception-master\doc\results.png (521261, 2020-08-11)
pytorch-deeplab-xception-master\LICENSE (1063, 2020-08-11)
pytorch-deeplab-xception-master\modeling (0, 2020-08-11)
pytorch-deeplab-xception-master\modeling\__init__.py (0, 2020-08-11)
pytorch-deeplab-xception-master\modeling\aspp.py (3602, 2020-08-11)
pytorch-deeplab-xception-master\modeling\backbone (0, 2020-08-11)
pytorch-deeplab-xception-master\modeling\backbone\__init__.py (514, 2020-08-11)
pytorch-deeplab-xception-master\modeling\backbone\drn.py (14620, 2020-08-11)
pytorch-deeplab-xception-master\modeling\backbone\mobilenet.py (5390, 2020-08-11)
pytorch-deeplab-xception-master\modeling\backbone\resnet.py (6222, 2020-08-11)
pytorch-deeplab-xception-master\modeling\backbone\xception.py (11554, 2020-08-11)
pytorch-deeplab-xception-master\modeling\decoder.py (2272, 2020-08-11)
pytorch-deeplab-xception-master\modeling\deeplab.py (2982, 2020-08-11)
pytorch-deeplab-xception-master\modeling\sync_batchnorm (0, 2020-08-11)
pytorch-deeplab-xception-master\modeling\sync_batchnorm\__init__.py (447, 2020-08-11)
pytorch-deeplab-xception-master\modeling\sync_batchnorm\batchnorm.py (12932, 2020-08-11)
pytorch-deeplab-xception-master\modeling\sync_batchnorm\comm.py (4440, 2020-08-11)
pytorch-deeplab-xception-master\modeling\sync_batchnorm\replicate.py (3218, 2020-08-11)
pytorch-deeplab-xception-master\modeling\sync_batchnorm\unittest.py (834, 2020-08-11)
pytorch-deeplab-xception-master\mypath.py (637, 2020-08-11)
pytorch-deeplab-xception-master\train.py (13621, 2020-08-11)
pytorch-deeplab-xception-master\train_coco.sh (191, 2020-08-11)
pytorch-deeplab-xception-master\train_voc.sh (209, 2020-08-11)
... ...
# 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.
![Results](doc/results.png)
### 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/zhanghang1***9/PyTorch-Encoding)
[Synchronized-BatchNorm-PyTorch](https://github.com/vacancy/Synchronized-BatchNorm-PyTorch)
[drn](https://github.com/fyu/drn)
近期下载者:
相关文件:
收藏者: