segmentationblwx / cssegmentation Goto Github PK

View Code? Open in Web Editor NEW

25.0 1.0 2.0 385 KB

CSSegmentation: An Open Source Continual Semantic Segmentation Toolbox Based on PyTorch.

Home Page: https://cssegmentation.readthedocs.io/en/latest/

License: Apache License 2.0

Python 97.91% Shell 2.09%

benchmark pytorch semantic-segmentation continual-learning

cssegmentation's Introduction

Documents: https://cssegmentation.readthedocs.io/en/latest/

Introduction

CSSegmentation: An Open Source Continual Semantic Segmentation Toolbox Based on PyTorch. You can star this repository to keep track of the project if it's helpful for you, thank you for your support.

Major Features

High Performance

The performance of re-implemented CSS algorithms is better than or comparable to the original paper.
Modular Design and Unified Benchmark

Various CSS methods are unified into several specific modules. Benefiting from this design, CSSegmentation can integrate a great deal of popular and contemporary continual semantic segmentation frameworks and then, train and test them on unified benchmarks.
Fewer Dependencies

CSSegmentation tries its best to avoid introducing more dependencies when reproducing novel continual semantic segmentation approaches.

Benchmark and Model Zoo

Supported Encoder

Encoder	Model Zoo	Paper Link	Code Snippet
ResNet	click	CVPR 2016	click

Supported Decoder

Decoder	Model Zoo	Paper Link	Code Snippet
Deeplabv3	click	ArXiv 2017	click

Supported Runner

Runner	Model Zoo	Paper Link	Code Snippet
EWF	click (under developing)	CVPR 2023	click
UCD	click (under developing)	TPAMI 2022	click
RCIL	click (under developing)	CVPR 2022	click
REMINDER	click (under developing)	CVPR 2022	click
CAF	click (under developing)	TMM 2022	click
SDR	click (under developing)	CVPR 2021	click
PLOP	click	CVPR 2021	click
MIB	click	CVPR 2020	click
ILT	click	ICCVW 2019	click

Supported Datasets

Dataset	Project Link	Paper Link	Code Snippet
ADE20k	Click	CVPR 2017	Click
PASCAL VOC	Click	IJCV 2010	Click

Citation

If you use this framework in your research, please cite this project:

@misc{csseg2023,
    author = {Zhenchao Jin},
    title = {CSSegmentation: An Open Source Continual Semantic Segmentation Toolbox Based on PyTorch},
    year = {2023},
    publisher = {GitHub},
    journal = {GitHub repository},
    howpublished = {\url{https://github.com/SegmentationBLWX/cssegmentation}},
}

cssegmentation's People

Contributors

Stargazers

Watchers

Forkers

charlespikachu anny-anny

cssegmentation's Issues

All tensors must be on devices[0]: 0

Hello, I encountered an error during the reproduction process. Could you help me?
Training with your own dataset reports an error:
(ID2) ubuntu@s414g1:~/cssegmentation$ bash scripts/dist_train.sh 4 /home/ubuntu/cssegmentation/csseg/co
nfigs/mib/mib_r101iabnd16_aspp_512x512_tile2-1_overlap.py

Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.

Filtering Images: 100%|█████████████████████████████████████████████| 274/274 [00:00<00:00, 391.28it/s]
Filtering Images: 100%|███████████████████████████████████████████████| 39/39 [00:00<00:00, 411.89it/s]
Selected optimization level O1: Insert automatic casts around Pytorch functions and Tensor methods.

Defaults for this optimization level are:
enabled : True
opt_level : O1
cast_model_type : None
patch_torch_functions : True
keep_batchnorm_fp32 : None
master_weights : None
loss_scale : dynamic
Processing user overrides (additional kwargs that are not None)...
After processing overrides, optimization options are:
enabled : True
opt_level : O1
cast_model_type : None
patch_torch_functions : True
keep_batchnorm_fp32 : None
master_weights : None
loss_scale : dynamic
Warning: multi_tensor_applier fused unscale kernel is unavailable, possibly because apex was installed without --cuda_ext --cpp_ext. Using Python fallback. Original ImportError was: ModuleNotFoundError("No module named 'amp_C'")
Traceback (most recent call last):
File "/home/ubuntu/cssegmentation/csseg/train.py", line 62, in
trainer_client.start()
File "/home/ubuntu/cssegmentation/csseg/train.py", line 54, in start
runner_client = BuildRunner(mode='TRAIN', cmd_args=cmd_args, runner_cfg=runner_cfg_task)
File "/home/ubuntu/cssegmentation/csseg/modules/runners/builder.py", line 29, in build
return super().build(module_cfg)
File "/home/ubuntu/cssegmentation/csseg/modules/utils/modulebuilder.py", line 26, in build
module = self.REGISTERED_MODULESmodule_type
File "/home/ubuntu/cssegmentation/csseg/modules/runners/mib.py", line 18, in init
super(MIBRunner, self).init(
File "/home/ubuntu/cssegmentation/csseg/modules/runners/base.py", line 120, in init
self.segmentor = BuildDistributedModel(model=self.segmentor, model_cfg=parallel_cfg['model_cfg'])
File "/home/ubuntu/cssegmentation/csseg/modules/parallel/model.py", line 12, in BuildDistributedModel
return nn.parallel.DistributedDataParallel(model, **model_cfg)
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/distributed.py", line 448, in init
self._ddp_init_helper()
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/distributed.py", line 503, in _ddp_init_helper
Traceback (most recent call last):
File "/home/ubuntu/cssegmentation/csseg/train.py", line 62, in
self._module_copies = replicate(self.module, self.device_ids, detach=True)
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/replicate.py", line 91, in replicate
param_copies = _broadcast_coalesced_reshape(params, devices, detach)
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/replicate.py", line 67, in _broadcast_coalesced_reshape
return comm.broadcast_coalesced(tensors, devices)
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/comm.py", line 58, in broadcast_coalesced
return torch._C._broadcast_coalesced(tensors, devices, buffer_size)
RuntimeError: All tensors must be on devices[0]: 0
trainer_client.start()
File "/home/ubuntu/cssegmentation/csseg/train.py", line 54, in start
runner_client = BuildRunner(mode='TRAIN', cmd_args=cmd_args, runner_cfg=runner_cfg_task)
File "/home/ubuntu/cssegmentation/csseg/modules/runners/builder.py", line 29, in build
return super().build(module_cfg)
File "/home/ubuntu/cssegmentation/csseg/modules/utils/modulebuilder.py", line 26, in build
module = self.REGISTERED_MODULESmodule_type
File "/home/ubuntu/cssegmentation/csseg/modules/runners/mib.py", line 18, in init
super(MIBRunner, self).init(
File "/home/ubuntu/cssegmentation/csseg/modules/runners/base.py", line 120, in init
self.segmentor = BuildDistributedModel(model=self.segmentor, model_cfg=parallel_cfg['model_cfg'])
File "/home/ubuntu/cssegmentation/csseg/modules/parallel/model.py", line 12, in BuildDistributedModel
return nn.parallel.DistributedDataParallel(model, **model_cfg)
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/distributed.py", line 448, in init
self._ddp_init_helper()
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/distributed.py", line 503, in _ddp_init_helper
self._module_copies = replicate(self.module, self.device_ids, detach=True)
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/replicate.py", line 91, in replicate
param_copies = _broadcast_coalesced_reshape(params, devices, detach)
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/replicate.py", line 67, in _broadcast_coalesced_reshape
return comm.broadcast_coalesced(tensors, devices)
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/comm.py", line 58, in broadcast_coalesced
return torch._C._broadcast_coalesced(tensors, devices, buffer_size)
RuntimeError: All tensors must be on devices[0]: 0
Traceback (most recent call last):
File "/home/ubuntu/cssegmentation/csseg/train.py", line 62, in
trainer_client.start()
File "/home/ubuntu/cssegmentation/csseg/train.py", line 54, in start
runner_client = BuildRunner(mode='TRAIN', cmd_args=cmd_args, runner_cfg=runner_cfg_task)
File "/home/ubuntu/cssegmentation/csseg/modules/runners/builder.py", line 29, in build
return super().build(module_cfg)
File "/home/ubuntu/cssegmentation/csseg/modules/utils/modulebuilder.py", line 26, in build
module = self.REGISTERED_MODULESmodule_type
File "/home/ubuntu/cssegmentation/csseg/modules/runners/mib.py", line 18, in init
super(MIBRunner, self).init(
File "/home/ubuntu/cssegmentation/csseg/modules/runners/base.py", line 120, in init
self.segmentor = BuildDistributedModel(model=self.segmentor, model_cfg=parallel_cfg['model_cfg'])
File "/home/ubuntu/cssegmentation/csseg/modules/parallel/model.py", line 12, in BuildDistributedModel
return nn.parallel.DistributedDataParallel(model, **model_cfg)
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/distributed.py", line 448, in init
self._ddp_init_helper()
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/distributed.py", line 503, in _ddp_init_helper
self._module_copies = replicate(self.module, self.device_ids, detach=True)
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/replicate.py", line 91, in replicate
param_copies = _broadcast_coalesced_reshape(params, devices, detach)
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/replicate.py", line 67, in _broadcast_coalesced_reshape
return comm.broadcast_coalesced(tensors, devices)
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/nn/parallel/comm.py", line 58, in broadcast_coalesced
return torch._C._broadcast_coalesced(tensors, devices, buffer_size)
RuntimeError: All tensors must be on devices[0]: 0
Killing subprocess 1686028
Killing subprocess 1686029
Killing subprocess 1686030
Killing subprocess 1686031
Traceback (most recent call last):
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/runpy.py", line 197, in _run_module_as_main
return _run_code(code, main_globals, None,
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/runpy.py", line 87, in _run_code
exec(code, run_globals)
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/distributed/launch.py", line 340, in
main()
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/distributed/launch.py", line 326, in main
sigkill_handler(signal.SIGTERM, None) # not coming back
File "/home/ubuntu/miniconda3/envs/ID2/lib/python3.9/site-packages/torch/distributed/launch.py", line 301, in sigkill_handler
raise subprocess.CalledProcessError(returncode=last_return_code, cmd=cmd)
subprocess.CalledProcessError: Command '['/home/ubuntu/miniconda3/envs/ID2/bin/python', '-u', 'csseg/train.py', '--local_rank=3', '--nproc_per_node', '4', '--cfgfilepath', '/home/ubuntu/cssegmentation/csseg/configs/mib/mib_r101iabnd16_aspp_512x512_tile2-1_overlap.py']' returned non-zero exit status 1.

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.