admin/Seg_Data_Server
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

first commit

Browse Source
This commit is contained in:
admin
2026-05-20 15:05:35 +08:00
commit ac09b26253
2048 changed files with 189478 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

76
Seg_All_In_One_MMSeg/configs/swin/README.md Normal file
View File

@@ -0,0 +1,76 @@
# Swin Transformer
> [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://arxiv.org/abs/2103.14030)
## Introduction
<!-- [BACKBONE] -->
<a href="https://github.com/microsoft/Swin-Transformer">Official Repo</a>
<a href="https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/backbones/swin.py#L524">Code Snippet</a>
## Abstract
<!-- [ABSTRACT] -->
This paper presents a new vision Transformer, called Swin Transformer, that capably serves as a general-purpose backbone for computer vision. Challenges in adapting Transformer from language to vision arise from differences between the two domains, such as large variations in the scale of visual entities and the high resolution of pixels in images compared to words in text. To address these differences, we propose a hierarchical Transformer whose representation is computed with Shifted windows. The shifted windowing scheme brings greater efficiency by limiting self-attention computation to non-overlapping local windows while also allowing for cross-window connection. This hierarchical architecture has the flexibility to model at various scales and has linear computational complexity with respect to image size. These qualities of Swin Transformer make it compatible with a broad range of vision tasks, including image classification (87.3 top-1 accuracy on ImageNet-1K) and dense prediction tasks such as object detection (58.7 box AP and 51.1 mask AP on COCO test-dev) and semantic segmentation (53.5 mIoU on ADE20K val). Its performance surpasses the previous state-of-the-art by a large margin of +2.7 box AP and +2.6 mask AP on COCO, and +3.2 mIoU on ADE20K, demonstrating the potential of Transformer-based models as vision backbones. The hierarchical design and the shifted window approach also prove beneficial for all-MLP architectures. The code and models are publicly available at [this https URL](https://github.com/microsoft/Swin-Transformer).
<!-- [IMAGE] -->
<div align=center>
<img src="https://user-images.githubusercontent.com/24582831/142902882-3fb9014c-11b6-47e9-aa14-500dfe7cbb1c.png" width="80%"/>
</div>
## Usage
We have provided pretrained models converted from [official repo](https://github.com/microsoft/Swin-Transformer)．
If you want to convert keys on your own to use official repositories' pre-trained models, we also provide a script [`swin2mmseg.py`](../../tools/model_converters/swin2mmseg.py) in the tools directory to convert the key of models from [the official repo](https://github.com/SwinTransformer/Swin-Transformer-Semantic-Segmentation) to MMSegmentation style.
```shell
python tools/model_converters/swin2mmseg.py ${PRETRAIN_PATH} ${STORE_PATH}
```
E.g.
```shell
python tools/model_converters/swin2mmseg.py https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_base_patch4_window7_224.pth pretrain/swin_base_patch4_window7_224.pth
```
This script convert model from `PRETRAIN_PATH` and store the converted model in `STORE_PATH`.
In our default setting, pretrained models and their corresponding [original models](https://github.com/microsoft/Swin-Transforme) models could be defined below:
| pretrained models | original models |
| ---------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------- |
| pretrain/swin_tiny_patch4_window7_224.pth | [swin_tiny_patch4_window7_224.pth](https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_tiny_patch4_window7_224.pth) |
| pretrain/swin_small_patch4_window7_224.pth | [swin_small_patch4_window7_224.pth](https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_small_patch4_window7_224.pth) |
| pretrain/swin_base_patch4_window7_224.pth | [swin_base_patch4_window7_224.pth](https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_base_patch4_window7_224.pth) |
| pretrain/swin_base_patch4_window7_224_22k.pth | [swin_base_patch4_window7_224_22k.pth](https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_base_patch4_window7_224_22k.pth) |
| pretrain/swin_base_patch4_window12_384.pth | [swin_base_patch4_window12_384.pth](https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_base_patch4_window12_384.pth) |
| pretrain/swin_base_patch4_window12_384_22k.pth | [swin_base_patch4_window12_384_22k.pth](https://github.com/SwinTransformer/storage/releases/download/v1.0.0/swin_base_patch4_window12_384_22k.pth) |
## Results and models
### ADE20K
| Method | Backbone | Crop Size | pretrain | pretrain img size | Batch Size | Lr schd | Mem (GB) | Inf time (fps) | Device | mIoU | mIoU(ms+flip) | config | download |
| ------- | -------- | --------- | ------------ | ----------------- | ---------- | ------- | -------- | -------------- | ------ | ----- | ------------: | ------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| UPerNet | Swin-T | 512x512 | ImageNet-1K | 224x224 | 16 | 160000 | 5.02 | 21.06 | V100 | 44.41 | 45.79 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/swin/swin-tiny-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_tiny_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_tiny_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210531_112542-e380ad3e.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_tiny_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_tiny_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210531_112542.log.json) |
| UPerNet | Swin-S | 512x512 | ImageNet-1K | 224x224 | 16 | 160000 | 6.17 | 14.72 | V100 | 47.72 | 49.24 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/swin/swin-small-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_small_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_small_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210526_192015-ee2fff1c.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_small_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_small_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210526_192015.log.json) |
| UPerNet | Swin-B | 512x512 | ImageNet-1K | 224x224 | 16 | 160000 | 7.61 | 12.65 | V100 | 47.99 | 49.57 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/swin/swin-base-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210526_192340-593b0e13.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210526_192340.log.json) |
| UPerNet | Swin-B | 512x512 | ImageNet-22K | 224x224 | 16 | 160000 | - | - | V100 | 50.13 | 51.9 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/swin/swin-base-patch4-window7-in22k-pre_upernet_8xb2-160k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_22K/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_22K_20210526_211650-762e2178.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_22K/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_22K_20210526_211650.log.json) |
| UPerNet | Swin-B | 512x512 | ImageNet-1K | 384x384 | 16 | 160000 | 8.52 | 12.10 | V100 | 48.35 | 49.65 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/swin/swin-base-patch4-window12-in1k-384x384-pre_upernet_8xb2-160k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_1K/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_1K_20210531_132020-05b22ea4.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_1K/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_1K_20210531_132020.log.json) |
| UPerNet | Swin-B | 512x512 | ImageNet-22K | 384x384 | 16 | 160000 | - | - | V100 | 50.76 | 52.4 | [config](https://github.com/open-mmlab/mmsegmentation/blob/main/configs/swin/swin-base-patch4-window12-in22k-384x384-pre_upernet_8xb2-160k_ade20k-512x512.py) | [model](https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_22K/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_22K_20210531_125459-429057bf.pth) \| [log](https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_22K/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_22K_20210531_125459.log.json) |
## Citation
```bibtex
@article{liu2021Swin,
title={Swin Transformer: Hierarchical Vision Transformer using Shifted Windows},
author={Liu, Ze and Lin, Yutong and Cao, Yue and Hu, Han and Wei, Yixuan and Zhang, Zheng and Lin, Stephen and Guo, Baining},
journal={arXiv preprint arXiv:2103.14030},
year={2021}
}
```

143
Seg_All_In_One_MMSeg/configs/swin/metafile.yaml Normal file
View File

@@ -0,0 +1,143 @@
Models:
- Name: swin-tiny-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512
In Collection: UPerNet
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 44.41
mIoU(ms+flip): 45.79
Config: configs/swin/swin-tiny-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- Swin-T
- UPerNet
Training Resources: 8x V100 GPUS
Memory (GB): 5.02
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_tiny_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_tiny_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210531_112542-e380ad3e.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_tiny_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_tiny_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210531_112542.log.json
Paper:
Title: 'Swin Transformer: Hierarchical Vision Transformer using Shifted Windows'
URL: https://arxiv.org/abs/2103.14030
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/backbones/swin.py#L524
Framework: PyTorch
- Name: swin-small-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512
In Collection: UPerNet
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 47.72
mIoU(ms+flip): 49.24
Config: configs/swin/swin-small-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- Swin-S
- UPerNet
Training Resources: 8x V100 GPUS
Memory (GB): 6.17
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_small_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_small_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210526_192015-ee2fff1c.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_small_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_small_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210526_192015.log.json
Paper:
Title: 'Swin Transformer: Hierarchical Vision Transformer using Shifted Windows'
URL: https://arxiv.org/abs/2103.14030
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/backbones/swin.py#L524
Framework: PyTorch
- Name: swin-base-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512
In Collection: UPerNet
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 47.99
mIoU(ms+flip): 49.57
Config: configs/swin/swin-base-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- Swin-B
- UPerNet
Training Resources: 8x V100 GPUS
Memory (GB): 7.61
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210526_192340-593b0e13.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210526_192340.log.json
Paper:
Title: 'Swin Transformer: Hierarchical Vision Transformer using Shifted Windows'
URL: https://arxiv.org/abs/2103.14030
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/backbones/swin.py#L524
Framework: PyTorch
- Name: swin-base-patch4-window7-in22k-pre_upernet_8xb2-160k_ade20k-512x512
In Collection: UPerNet
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 50.13
mIoU(ms+flip): 51.9
Config: configs/swin/swin-base-patch4-window7-in22k-pre_upernet_8xb2-160k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- Swin-B
- UPerNet
Training Resources: 8x V100 GPUS
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_22K/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_22K_20210526_211650-762e2178.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_22K/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_22K_20210526_211650.log.json
Paper:
Title: 'Swin Transformer: Hierarchical Vision Transformer using Shifted Windows'
URL: https://arxiv.org/abs/2103.14030
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/backbones/swin.py#L524
Framework: PyTorch
- Name: swin-base-patch4-window12-in1k-384x384-pre_upernet_8xb2-160k_ade20k-512x512
In Collection: UPerNet
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 48.35
mIoU(ms+flip): 49.65
Config: configs/swin/swin-base-patch4-window12-in1k-384x384-pre_upernet_8xb2-160k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- Swin-B
- UPerNet
Training Resources: 8x V100 GPUS
Memory (GB): 8.52
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_1K/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_1K_20210531_132020-05b22ea4.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_1K/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_1K_20210531_132020.log.json
Paper:
Title: 'Swin Transformer: Hierarchical Vision Transformer using Shifted Windows'
URL: https://arxiv.org/abs/2103.14030
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/backbones/swin.py#L524
Framework: PyTorch
- Name: swin-base-patch4-window12-in22k-384x384-pre_upernet_8xb2-160k_ade20k-512x512
In Collection: UPerNet
Results:
Task: Semantic Segmentation
Dataset: ADE20K
Metrics:
mIoU: 50.76
mIoU(ms+flip): 52.4
Config: configs/swin/swin-base-patch4-window12-in22k-384x384-pre_upernet_8xb2-160k_ade20k-512x512.py
Metadata:
Training Data: ADE20K
Batch Size: 16
Architecture:
- Swin-B
- UPerNet
Training Resources: 8x V100 GPUS
Weights: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_22K/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_22K_20210531_125459-429057bf.pth
Training log: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_22K/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_22K_20210531_125459.log.json
Paper:
Title: 'Swin Transformer: Hierarchical Vision Transformer using Shifted Windows'
URL: https://arxiv.org/abs/2103.14030
Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/backbones/swin.py#L524
Framework: PyTorch

14
Seg_All_In_One_MMSeg/configs/swin/swin-base-patch4-window12-in1k-384x384-pre_upernet_8xb2-160k_ade20k-512x512.py Normal file
View File

@@ -0,0 +1,14 @@
_base_ = [
'swin-tiny-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py'
]
checkpoint_file = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/swin/swin_base_patch4_window12_384_20220317-55b0104a.pth' # noqa
model = dict(
backbone=dict(
init_cfg=dict(type='Pretrained', checkpoint=checkpoint_file),
pretrain_img_size=384,
embed_dims=128,
depths=[2, 2, 18, 2],
num_heads=[4, 8, 16, 32],
window_size=12),
decode_head=dict(in_channels=[128, 256, 512, 1024], num_classes=150),
auxiliary_head=dict(in_channels=512, num_classes=150))

7
Seg_All_In_One_MMSeg/configs/swin/swin-base-patch4-window12-in22k-384x384-pre_upernet_8xb2-160k_ade20k-512x512.py Normal file
View File

@@ -0,0 +1,7 @@
_base_ = [
'./swin-base-patch4-window12-in1k-384x384-pre_upernet_8xb2-160k_ade20k-512x512.py' # noqa
]
checkpoint_file = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/swin/swin_base_patch4_window12_384_22k_20220317-e5c09f74.pth' # noqa
model = dict(
backbone=dict(
init_cfg=dict(type='Pretrained', checkpoint=checkpoint_file)))

12
Seg_All_In_One_MMSeg/configs/swin/swin-base-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py Normal file
View File

@@ -0,0 +1,12 @@
_base_ = [
'./swin-tiny-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py'
]
checkpoint_file = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/swin/swin_base_patch4_window7_224_20220317-e9b98025.pth' # noqa
model = dict(
backbone=dict(
init_cfg=dict(type='Pretrained', checkpoint=checkpoint_file),
embed_dims=128,
depths=[2, 2, 18, 2],
num_heads=[4, 8, 16, 32]),
decode_head=dict(in_channels=[128, 256, 512, 1024], num_classes=150),
auxiliary_head=dict(in_channels=512, num_classes=150))

7
Seg_All_In_One_MMSeg/configs/swin/swin-base-patch4-window7-in22k-pre_upernet_8xb2-160k_ade20k-512x512.py Normal file
View File

@@ -0,0 +1,7 @@
_base_ = [
'./swin-base-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py'
]
checkpoint_file = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/swin/swin_base_patch4_window7_224_22k_20220317-4f79f7c0.pth' # noqa
model = dict(
backbone=dict(
init_cfg=dict(type='Pretrained', checkpoint=checkpoint_file)))

10
Seg_All_In_One_MMSeg/configs/swin/swin-large-patch4-window12-in22k-384x384-pre_upernet_8xb2-160k_ade20k-512x512.py Normal file
View File

@@ -0,0 +1,10 @@
_base_ = [
'swin-large-patch4-window7-in22k-pre_upernet_'
'8xb2-160k_ade20k-512x512.py'
]
checkpoint_file = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/swin/swin_large_patch4_window12_384_22k_20220412-6580f57d.pth' # noqa
model = dict(
backbone=dict(
init_cfg=dict(type='Pretrained', checkpoint=checkpoint_file),
pretrain_img_size=384,
window_size=12))

15
Seg_All_In_One_MMSeg/configs/swin/swin-large-patch4-window7-in22k-pre_upernet_8xb2-160k_ade20k-512x512.py Normal file
View File

@@ -0,0 +1,15 @@
_base_ = [
'swin-tiny-patch4-window7-in1k-pre_upernet_8xb2-160k_'
'ade20k-512x512.py'
]
checkpoint_file = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/swin/swin_large_patch4_window7_224_22k_20220412-aeecf2aa.pth' # noqa
model = dict(
backbone=dict(
init_cfg=dict(type='Pretrained', checkpoint=checkpoint_file),
pretrain_img_size=224,
embed_dims=192,
depths=[2, 2, 18, 2],
num_heads=[6, 12, 24, 48],
window_size=7),
decode_head=dict(in_channels=[192, 384, 768, 1536], num_classes=150),
auxiliary_head=dict(in_channels=768, num_classes=150))

10
Seg_All_In_One_MMSeg/configs/swin/swin-small-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py Normal file
View File

@@ -0,0 +1,10 @@
_base_ = [
'./swin-tiny-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py'
]
checkpoint_file = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/swin/swin_small_patch4_window7_224_20220317-7ba6d6dd.pth' # noqa
model = dict(
backbone=dict(
init_cfg=dict(type='Pretrained', checkpoint=checkpoint_file),
depths=[2, 2, 18, 2]),
decode_head=dict(in_channels=[96, 192, 384, 768], num_classes=150),
auxiliary_head=dict(in_channels=384, num_classes=150))

52
Seg_All_In_One_MMSeg/configs/swin/swin-tiny-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py Normal file
View File

@@ -0,0 +1,52 @@
_base_ = [
'../_base_/models/upernet_swin.py', '../_base_/datasets/ade20k.py',
'../_base_/default_runtime.py', '../_base_/schedules/schedule_160k.py'
]
crop_size = (512, 512)
data_preprocessor = dict(size=crop_size)
checkpoint_file = 'https://download.openmmlab.com/mmsegmentation/v0.5/pretrain/swin/swin_tiny_patch4_window7_224_20220317-1cdeb081.pth' # noqa
model = dict(
data_preprocessor=data_preprocessor,
backbone=dict(
init_cfg=dict(type='Pretrained', checkpoint=checkpoint_file),
embed_dims=96,
depths=[2, 2, 6, 2],
num_heads=[3, 6, 12, 24],
window_size=7,
use_abs_pos_embed=False,
drop_path_rate=0.3,
patch_norm=True),
decode_head=dict(in_channels=[96, 192, 384, 768], num_classes=150),
auxiliary_head=dict(in_channels=384, num_classes=150))
# AdamW optimizer, no weight decay for position embedding & layer norm
# in backbone
optim_wrapper = dict(
_delete_=True,
type='OptimWrapper',
optimizer=dict(
type='AdamW', lr=0.00006, betas=(0.9, 0.999), weight_decay=0.01),
paramwise_cfg=dict(
custom_keys={
'absolute_pos_embed': dict(decay_mult=0.),
'relative_position_bias_table': dict(decay_mult=0.),
'norm': dict(decay_mult=0.)
}))
param_scheduler = [
dict(
type='LinearLR', start_factor=1e-6, by_epoch=False, begin=0, end=1500),
dict(
type='PolyLR',
eta_min=0.0,
power=1.0,
begin=1500,
end=160000,
by_epoch=False,
)
]
# By default, models are trained on 8 GPUs with 2 images per GPU
train_dataloader = dict(batch_size=2)
val_dataloader = dict(batch_size=1)
test_dataloader = val_dataloader

57
Seg_All_In_One_MMSeg/configs/swin/swin-tiny-patch4-window7_upernet_1xb8-20k_levir-256x256.py Normal file
View File

@@ -0,0 +1,57 @@
_base_ = [
'../_base_/models/upernet_swin.py', '../_base_/datasets/levir_256x256.py',
'../_base_/default_runtime.py', '../_base_/schedules/schedule_20k.py'
]
crop_size = (256, 256)
norm_cfg = dict(type='BN', requires_grad=True)
data_preprocessor = dict(
size=crop_size,
type='SegDataPreProcessor',
mean=[123.675, 116.28, 103.53, 123.675, 116.28, 103.53],
std=[58.395, 57.12, 57.375, 58.395, 57.12, 57.375],
bgr_to_rgb=False)
model = dict(
data_preprocessor=data_preprocessor,
backbone=dict(
in_channels=6,
embed_dims=96,
depths=[2, 2, 6, 2],
num_heads=[3, 6, 12, 24],
window_size=7,
use_abs_pos_embed=False,
drop_path_rate=0.3,
patch_norm=True),
decode_head=dict(in_channels=[96, 192, 384, 768], num_classes=2),
auxiliary_head=dict(in_channels=384, num_classes=2))
# AdamW optimizer, no weight decay for position embedding & layer norm
# in backbone
optim_wrapper = dict(
_delete_=True,
type='OptimWrapper',
optimizer=dict(
type='AdamW', lr=0.00006, betas=(0.9, 0.999), weight_decay=0.01),
paramwise_cfg=dict(
custom_keys={
'absolute_pos_embed': dict(decay_mult=0.),
'relative_position_bias_table': dict(decay_mult=0.),
'norm': dict(decay_mult=0.)
}))
param_scheduler = [
dict(
type='LinearLR', start_factor=1e-6, by_epoch=False, begin=0, end=1500),
dict(
type='PolyLR',
eta_min=0.0,
power=1.0,
begin=1500,
end=20000,
by_epoch=False,
)
]
train_dataloader = dict(batch_size=4)
val_dataloader = dict(batch_size=1)
test_dataloader = val_dataloader