forked from DoctorKey/C-3PO
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
0 parents
commit 0c3df07
Showing
153 changed files
with
6,261 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,3 @@ | ||
__pycache__ | ||
*.ipynb | ||
.ipynb_checkpoints |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,92 @@ | ||
# C-3PO | ||
|
||
This is the official Pytorch implementation for [How to Reduce Change Detection to Semantic Segmentation](https://arxiv.org/abs/2206.07557). | ||
|
||
Overall, we present a new paradigm that reduces change detection to semantic segmentation which means tailoring an existing and powerful semantic segmentation network to solve change detection. | ||
|
||
![reduce](imgs/reduce.jpg "reduce") | ||
|
||
Our analysis suggests that there are three possible change types within the change detection task and they should be learned separately. Therefore, we propose a MTF (Merge Temporal Features ) module to learn these changes. | ||
|
||
![changes](imgs/changes.jpg "changes") | ||
|
||
We propose a simple but effective network, called C-3PO (Combine 3 POssible change types), detects changes in pixel-level, and can be considered as a new baseline network in this field. | ||
|
||
![C-3PO](imgs/C3PO.jpg "C-3PO") | ||
|
||
![MTF](imgs/MTF.jpg "MTF") | ||
|
||
![MSF](imgs/MSF.jpg "MSF") | ||
|
||
## Requirements | ||
|
||
* Python3 | ||
* PyTorch | ||
* Torchvision | ||
* pycocotools | ||
* timm | ||
|
||
`Python3`, `Pytorch` and `Torchvision` are necessary. `pycocotools` is required for the `COCO` dataset. `timm` is required for the `Swin Transformer` backbone. | ||
|
||
If you want to use `CSCDNet` in our project, please follow their [instructions](https://github.com/kensakurada/sscdnet) to install the `correlation` module. | ||
|
||
## Prepare the dataset | ||
|
||
There are three datasets needed in this projects: | ||
* COCO | ||
* PCD | ||
* VL-CMU-CD | ||
* ChangeSim | ||
|
||
Please refer to `src/dataset/path_config.py` to understand the folder structure of each dataset. And edit `data_path` according to your system. | ||
|
||
Please follow this [site](https://kensakurada.github.io/pcd_dataset.html) to download the PCD dataset. You may need to send e-mails to Takayuki Okatani. | ||
|
||
You can download VL-CMU-CD by this [link](https://drive.google.com/file/d/0B-IG2NONFdciOWY5QkQ3OUgwejQ/view?resourcekey=0-rEzCjPFmDFjt4UMWamV4Eg). | ||
|
||
Please follow this [page](https://github.com/SAMMiCA/ChangeSim) to prepare the ChangeSim dataset. | ||
|
||
## Run | ||
|
||
**training** | ||
``` | ||
python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --model resnet18_mtf_msf_fcn --mtf id --msf 4 --warmup --loss-weight | ||
``` | ||
|
||
**testing** | ||
``` | ||
python3 src/train.py --test-only --model resnet18_mtf_msf_fcn --mtf id --msf 4 --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --resume [checkpoint.pth] | ||
``` | ||
|
||
We provide all shells to reproduce the results in our paper. Please check files in the `exp` folder. You can use below commands to run experiments. | ||
|
||
``` | ||
source exp/sota/resnet18_mtf_id_msf4_deeplabv3_cmu.sh | ||
train | ||
``` | ||
|
||
## Visualization | ||
|
||
![CMU](imgs/CMU.png "CMU") | ||
|
||
## Citation | ||
|
||
If you find the work useful for your research, please cite: | ||
|
||
``` | ||
@article{wang2022c3po, | ||
title={How to Reduce Change Detection to Semantic Segmentation}, | ||
author={Wang, Guo-Hua and Gao, Bin-Bin and Wang, Chengjie}, | ||
journal={Pattern Recognition}, | ||
year={2023} | ||
} | ||
``` | ||
|
||
## reference | ||
|
||
* https://github.com/pytorch/vision/tree/main/references/segmentation | ||
* https://github.com/kensakurada/sscdnet | ||
* https://github.com/rcdaudt/fully_convolutional_change_detection | ||
* https://github.com/leonardoaraujosantos/ChangeNet | ||
* https://github.com/SAMMiCA/ChangeSim | ||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV_woRot --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 0 --input-size 256 --randomflip 0 --randomcrop --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 0 --input-size 256 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV_woRot --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 0 --input-size 256 --randomflip 0.5 --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 0 --input-size 256 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV_woRot --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 0 --input-size 256 --randomflip 0.5 --randomcrop --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 0 --input-size 256 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV_woRot --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 0 --input-size 256 --randomflip 0 --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 0 --input-size 256 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 0 --input-size 256 --randomflip 0 --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 0 --input-size 256 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 0 --input-size 256 --randomflip 0 --randomcrop --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 0 --input-size 256 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 0 --input-size 256 --randomflip 0.5 --randomcrop --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf iade --msf 4 --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 0 --input-size 256 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD_Raw --test-dataset VL_CMU_CD_Raw --input-size 512 --randomflip 0 --randomcrop --model resnet18_mtf_msf_fcn --mtf id --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf id --msf 4 --train-dataset VL_CMU_CD_Raw --test-dataset VL_CMU_CD_Raw --input-size 512 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD_Raw --test-dataset VL_CMU_CD_Raw --input-size 512 --randomflip 0.5 --model resnet18_mtf_msf_fcn --mtf id --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf id --msf 4 --train-dataset VL_CMU_CD_Raw --test-dataset VL_CMU_CD_Raw --input-size 512 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD_Raw --test-dataset VL_CMU_CD_Raw --input-size 512 --randomflip 0.5 --randomcrop --model resnet18_mtf_msf_fcn --mtf id --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf id --msf 4 --train-dataset VL_CMU_CD_Raw --test-dataset VL_CMU_CD_Raw --input-size 512 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD_Raw --test-dataset VL_CMU_CD_Raw --input-size 512 --randomflip 0 --model resnet18_mtf_msf_fcn --mtf id --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf id --msf 4 --train-dataset VL_CMU_CD_Raw --test-dataset VL_CMU_CD_Raw --input-size 512 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --randomflip 0 --model resnet18_mtf_msf_fcn --mtf id --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf id --msf 4 --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --randomflip 0 --randomcrop --model resnet18_mtf_msf_fcn --mtf id --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf id --msf 4 --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --randomflip 0.5 --randomcrop --model resnet18_mtf_msf_fcn --mtf id --msf 4 --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_fcn --mtf id --msf 4 --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --resume output/deeplabv3_bi_fpn4_bdae_vgg16bn_PCD_CV_0_256/2021-10-02_10:58:34/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,5 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --model resnet18_mtf_msf_deeplabv3 --mtf a --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --model resnet18_mtf_msf_deeplabv3 --mtf a --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --resume output/deeplabv3_tri_mulitfeature_resnet18_VL_CMU_CD_0_512/2021-09-15_09:59:00/checkpoint.pth" |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 4 --input-size 256 --model resnet18_mtf_msf_deeplabv3 --mtf a --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_deeplabv3 --mtf a --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 4 --input-size 256 --resume output/fcn_tri_fpn4_resnet18_PCD_CV_0_256/2021-09-18_09:01:41/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,5 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --model resnet18_mtf_msf_deeplabv3 --mtf d --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --model resnet18_mtf_msf_deeplabv3 --mtf d --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --resume output/deeplabv3_tri_mulitfeature_resnet18_VL_CMU_CD_0_512/2021-09-15_09:59:00/checkpoint.pth" |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 4 --input-size 256 --model resnet18_mtf_msf_deeplabv3 --mtf d --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_deeplabv3 --mtf d --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 4 --input-size 256 --resume output/fcn_tri_fpn4_resnet18_PCD_CV_0_256/2021-09-18_09:01:41/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,5 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --model resnet18_mtf_msf_deeplabv3 --mtf e --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --model resnet18_mtf_msf_deeplabv3 --mtf e --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --resume output/deeplabv3_tri_mulitfeature_resnet18_VL_CMU_CD_0_512/2021-09-15_09:59:00/checkpoint.pth" |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 4 --input-size 256 --model resnet18_mtf_msf_deeplabv3 --mtf e --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_deeplabv3 --mtf e --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 4 --input-size 256 --resume output/fcn_tri_fpn4_resnet18_PCD_CV_0_256/2021-09-18_09:01:41/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,5 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --model resnet18_mtf_msf_deeplabv3 --mtf i --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --model resnet18_mtf_msf_deeplabv3 --mtf i --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --resume output/deeplabv3_tri_mulitfeature_resnet18_VL_CMU_CD_0_512/2021-09-15_09:59:00/checkpoint.pth" |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 4 --input-size 256 --model resnet18_mtf_msf_deeplabv3 --mtf i --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_deeplabv3 --mtf i --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 4 --input-size 256 --resume output/fcn_tri_fpn4_resnet18_PCD_CV_0_256/2021-09-18_09:01:41/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,5 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --model resnet18_mtf_msf_deeplabv3 --mtf ia --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --model resnet18_mtf_msf_deeplabv3 --mtf ia --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --resume output/deeplabv3_tri_mulitfeature_resnet18_VL_CMU_CD_0_512/2021-09-15_09:59:00/checkpoint.pth" |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 4 --input-size 256 --model resnet18_mtf_msf_deeplabv3 --mtf ia --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_deeplabv3 --mtf ia --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 4 --input-size 256 --resume output/fcn_tri_fpn4_resnet18_PCD_CV_0_256/2021-09-18_09:01:41/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,5 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --model resnet18_mtf_msf_deeplabv3 --mtf iad --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --model resnet18_mtf_msf_deeplabv3 --mtf iad --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --resume output/deeplabv3_tri_mulitfeature_resnet18_VL_CMU_CD_0_512/2021-09-15_09:59:00/checkpoint.pth" |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 4 --input-size 256 --model resnet18_mtf_msf_deeplabv3 --mtf iad --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_deeplabv3 --mtf iad --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 4 --input-size 256 --resume output/fcn_tri_fpn4_resnet18_PCD_CV_0_256/2021-09-18_09:01:41/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,5 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --model resnet18_mtf_msf_deeplabv3 --mtf iade --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --model resnet18_mtf_msf_deeplabv3 --mtf iade --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --resume output/deeplabv3_tri_mulitfeature_resnet18_VL_CMU_CD_0_512/2021-09-15_09:59:00/checkpoint.pth" |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 4 --input-size 256 --model resnet18_mtf_msf_deeplabv3 --mtf iade --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_deeplabv3 --mtf iade --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 4 --input-size 256 --resume output/fcn_tri_fpn4_resnet18_PCD_CV_0_256/2021-09-18_09:01:41/best.pth" | ||
|
||
|
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,5 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --model resnet18_mtf_msf_deeplabv3 --mtf iae --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --model resnet18_mtf_msf_deeplabv3 --mtf iae --train-dataset VL_CMU_CD --test-dataset VL_CMU_CD --input-size 512 --resume output/deeplabv3_tri_mulitfeature_resnet18_VL_CMU_CD_0_512/2021-09-15_09:59:00/checkpoint.pth" |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,7 @@ | ||
#!/bin/bash | ||
|
||
train="python3 -m torch.distributed.launch --nproc_per_node=4 --use_env src/train.py --train-dataset PCD_CV --test-dataset GSV --test-dataset2 TSUNAMI --data-cv 4 --input-size 256 --model resnet18_mtf_msf_deeplabv3 --mtf iae --warmup --loss bi --loss-weight" | ||
|
||
test="python3 src/train.py --test-only --save-imgs --model resnet18_mtf_msf_deeplabv3 --mtf iae --train-dataset PCD_CV --test-dataset TSUNAMI --data-cv 4 --input-size 256 --resume output/fcn_tri_fpn4_resnet18_PCD_CV_0_256/2021-09-18_09:01:41/best.pth" | ||
|
||
|
Oops, something went wrong.