Official implementation of our CVPR2021 paper "OTA: Optimal Transport Assignment for Object Detection" in Pytorch.

Last update: Jan 3, 2023

Related tags

Deep Learning OTA

Overview

OTA: Optimal Transport Assignment for Object Detection

This project provides an implementation for our CVPR2021 paper "OTA: Optimal Transport Assignment for Object Detection" on PyTorch.

Requirements

cvpods

Get Started

install cvpods locally (requires cuda to compile)

python3 -m pip install 'git+https://github.com/Megvii-BaseDetection/cvpods.git'
# (add --user if you don't have permission)

# Or, to install it from a local clone:
git clone https://github.com/Megvii-BaseDetection/cvpods.git
python3 -m pip install -e cvpods

# Or,
pip install -r requirements.txt
python3 setup.py build develop

prepare datasets

cd /path/to/cvpods/datasets
ln -s /path/to/your/coco/dataset coco

Train & Test

git clone https://github.com/Megvii-BaseDetection/OTA.git
cd playground/detection/coco/ota.res50.fpn.coco.800size.1x  # for example

# Train
pods_train --num-gpus 8

# Test
pods_test --num-gpus 8 \
    MODEL.WEIGHTS /path/to/your/save_dir/ckpt.pth # optional
    OUTPUT_DIR /path/to/your/save_dir # optional

# Multi node training
## sudo apt install net-tools ifconfig
pods_train --num-gpus 8 --num-machines N --machine-rank 0/1/.../N-1 --dist-url "tcp://MASTER_IP:port"

Results on COCO val set

Model	Backbone	LR Sched.	mAP	Recall	AP50/AP75/APs/APm/APl	Download
RetinaNet	R50	1x	36.5	53.4	56.2/39.3/21.9/40.5/47.7	-
Faster R-CNN	R50	1x	38.1	52.2	58.9/41.0/22.5/41.5/48.9	-
FCOS	R50	1x	38.7	57.0	57.5/41.7/22.6/42.7/49.9	-
FreeAnchor	R50	1x	38.4	55.4	57.0/41.1/21.9/41.7/51.8	-
ATSS	R50	1x	39.4	57.7	57.5/42.7/22.9/42.9/51.2	-
PAA(w/. Voting)	R50	1x	40.4	-	-	-
OTA	R50	1x	40.7	59.0	58.4/44.3/23.2/45.0/53.6	weights

Results on COCO test-dev

Model	Backbone	LR Sched.	Training Scale (ShortSide)	mAP	AP50/AP75/APs/APm/APl	Download
OTA	R101	2x	640~800	45.3	63.5/49.3/26.9/48.8/56.1	weights
OTA	X101	2x	640~800	47.0	65.8/51.1/29.2/50.4/57.9	weights
OTA	X101-DCN	2x	640~800	49.2	67.6/53.5/30.0/52.5/62.3	weights
OTA*	X101-DCN	2x	640~800	51.5	68.6/57.1/34.1/53.7/64.1	weights

* stands for ATSS-style testing time augmentation. To enable testing time augmentation, add/modify the following code frac in the corresponding config.py

TEST=dict(
    DETECTIONS_PER_IMAGE=300,
    AUG=dict(
        ENABLED=True,
        MAX_SIZE=3000,
        MIN_SIZES=(400, 500, 600, 640, 700, 900, 1000, 1100, 1200, 1300, 1400, 1800),
        EXTRA_SIZES=((800, 1333),),
        SCALE_FILTER=True,
        SCALE_RANGES=(
        [96, 10000], [96, 10000], [64, 10000], [64, 10000], [64, 10000], [0, 10000], [0, 10000], [0, 256], [0, 256], [0, 192], [0, 192], [0, 96], [0, 10000])
    )
),

Acknowledgement

This repo is developed based on cvpods. Please check cvpods for more details and features.

License

This repo is released under the Apache 2.0 license. Please see the LICENSE file for more information.

Comments

Reproduction

I changed nothing for the config file but only got 39.9 mAP when training res50 by myself. And I tested with your provided weight file and only got 31 mAP for res50 and 36 for res101.

opened by strongwolf 4
Training time for dynamic k estimation vs fixed k

Hi, thanks for your great work on OTA. I have tried to incorporate OTA into my model and found a huge difference in training time for dynamic k estimation vs fixed k settings. I am using 4 Tesla P100 to train. It takes about 2.5 days for fixed k while it takes more than 6 days to train for dynamic k estimation. I am wondering what your training time is like on both settings. Do you face similar issues? Thanks a lot!

opened by QinghangHong1 2
how to train with one GPU?

I used pods_train --num-gpus 1 , but I got error: Traceback (most recent call last): File "/home/***/anaconda3/envs/torch1.8/lib/python3.7/runpy.py", line 193, in _run_module_as_main "main", mod_spec) File "/home/***/anaconda3/envs/torch1.8/lib/python3.7/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/*/.vscode/extensions/ms-python.python-2021.5.926500501/pythonFiles/lib/python/debugpy/main.py", line 45, in cli.main() File "/home//.vscode/extensions/ms-python.python-2021.5.926500501/pythonFiles/lib/python/debugpy/../debugpy/server/cli.py", line 444, in main run() File "/home//.vscode/extensions/ms-python.python-2021.5.926500501/pythonFiles/lib/python/debugpy/../debugpy/server/cli.py", line 285, in run_file runpy.run_path(target_as_str, run_name=compat.force_str("main")) File "/home//anaconda3/envs/torch1.8/lib/python3.7/runpy.py", line 263, in run_path pkg_name=pkg_name, script_name=fname) File "/home//anaconda3/envs/torch1.8/lib/python3.7/runpy.py", line 96, in _run_module_code mod_name, mod_spec, pkg_name, script_name) File "/home//anaconda3/envs/torch1.8/lib/python3.7/runpy.py", line 85, in _run_code exec(code, run_globals) File "/media//D/05_source_code/01_torch/detection/cvpods/tools/train_net.py", line 124, in args=(args,), File "/media//D/05_source_code/01_torch/detection/cvpods/cvpods/engine/launch.py", line 56, in launch main_func(args) File "/media//D/05_source_code/01_torch/detection/cvpods/tools/train_net.py", line 110, in main runner.train() File "/media//D/05_source_code/01_torch/detection/cvpods/cvpods/engine/runner.py", line 271, in train super().train(self.start_iter, self.start_epoch, self.max_iter) File "/media//D/05_source_code/01_torch/detection/cvpods/cvpods/engine/base_runner.py", line 84, in train self.run_step() File "/media//D/05_source_code/01_torch/detection/cvpods/cvpods/engine/base_runner.py", line 185, in run_step loss_dict = self.model(data) File "/home//anaconda3/envs/torch1.8/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl result = self.forward(input, kwargs) File "/media//D/05_source_code/01_torch/detection/cvpods/playground/detection/coco/ota.res50.fpn.coco.800size.1x/fcos.py", line 218, in forward gt_ious, box_cls, box_delta, box_iou) File "/media//D/05_source_code/01_torch/detection/cvpods/playground/detection/coco/ota.res50.fpn.coco.800size.1x/fcos.py", line 383, in losses dist.all_reduce(num_foreground) File "/home//anaconda3/envs/torch1.8/lib/python3.7/site-packages/torch/distributed/distributed_c10d.py", line 1168, in all_reduce default_pg = _get_default_group() File "/home//anaconda3/envs/torch1.8/lib/python3.7/site-packages/torch/distributed/distributed_c10d.py", line 347, in _get_default_group raise RuntimeError("Default process group has not been initialized, " RuntimeError: Default process group has not been initialized, please make sure to call init_process_group.

how to train with one GPU? Thank you

opened by SidneyRey 1
Reason for using log scale sinkhorn algorithm

Hi Thanks for providing wonderful work! I have a question about the sinkhorn distances implementation.

In the forward function of sinkhorn distances (below), you use log scale updating rule which is not same as original algorithm, and add v_t,u_t to update v_t+1,u_t+1. Whats the reason for that? https://github.com/Megvii-BaseDetection/OTA/blob/2c85b4d0f9031396854aae969330dde2ab5eacbd/playground/detection/coco/ota.x101.dcnv2.fpn.coco.800size.1x/fcos.py#L693-L699

opened by jihwanp 0
Training Error

I met a unexpected error ,shown as : 2022-08-31 17:30:53 | ERROR | cvpods.engine.launch:64 - An error has been caught in function 'launch', process 'MainProcess' (1731), thread 'MainThread' (140323764021056): Traceback (most recent call last):

File "/root/cvpods/tools/train_net.py", line 157, in launch( └ <function launch at 0x7f9d54537040>

File "/root/cvpods/cvpods/engine/launch.py", line 64, in launch main_func(*args) │ └ (Namespace(clearml=False, dir='atss/atss.res50.fpn.coco.800size.1x/', dist_url='tcp://127.0.0.1:49152', eval_only=False, mach... └ <function main at 0x7f9d54553b80>

File "/root/cvpods/tools/train_net.py", line 89, in main runner.resume_or_load(resume=args.resume) │ │ │ └ False │ │ └ Namespace(clearml=False, dir='atss/atss.res50.fpn.coco.800size.1x/', dist_url='tcp://127.0.0.1:49152', eval_only=False, machi... │ └ <function DefaultRunner.resume_or_load at 0x7f9d545531f0> └ <cvpods.engine.runner.DefaultRunner object at 0x7f9d545a77c0>

File "/root/cvpods/cvpods/engine/runner.py", line 172, in resume_or_load self.start_iter = (self.checkpointer.resume_or_load( │ │ │ │ └ <function Checkpointer.resume_or_load at 0x7f9e113c78b0> │ │ │ └ <cvpods.checkpoint.checkpoint.DefaultCheckpointer object at 0x7f9cc6676730> │ │ └ <cvpods.engine.runner.DefaultRunner object at 0x7f9d545a77c0> │ └ 0 └ <cvpods.engine.runner.DefaultRunner object at 0x7f9d545a77c0>

File "/root/cvpods/cvpods/checkpoint/checkpoint.py", line 177, in resume_or_load return self.load(path) │ │ └ 'detectron2://ImageNetPretrained/MSRA/R-50.pkl' │ └ <function Checkpointer.load at 0x7f9e113c7670> └ <cvpods.checkpoint.checkpoint.DefaultCheckpointer object at 0x7f9cc6676730>

File "/root/cvpods/cvpods/checkpoint/checkpoint.py", line 112, in load assert megfile.smart_isfile(path), "Checkpoint {} not found!".format(path) │ │ │ └ 'detectron2://ImageNetPretrained/MSRA/R-50.pkl' │ │ └ 'detectron2://ImageNetPretrained/MSRA/R-50.pkl' │ └ <function smart_isfile at 0x7f9e114143a0> └ <module 'megfile' from '/usr/local/lib/python3.8/dist-packages/megfile/init.py'>

File "/usr/local/lib/python3.8/dist-packages/megfile/smart.py", line 101, in smart_isfile return SmartPath(path).is_file(followlinks=followlinks) │ │ └ False │ └ 'detectron2://ImageNetPretrained/MSRA/R-50.pkl' └ <class 'megfile.smart_path.SmartPath'>

File "/usr/local/lib/python3.8/dist-packages/megfile/smart_path.py", line 14, in smart_method return getattr(self.pathlike, name)(*args, **kwargs) │ │ │ │ └ {'followlinks': False} │ │ │ └ () │ │ └ 'is_file' │ └ Detectron2Path('ImageNetPretrained/MSRA/R-50.pkl') └ SmartPath('detectron2://ImageNetPretrained/MSRA/R-50.pkl')

TypeError: is_file() got an unexpected keyword argument 'followlinks'

opened by PanffeeReal 0
How to solve the problem:PermissionError: [Errno 13] Permission denied: '/data'

2022-03-15 15:37:57.573 | INFO | main::157 - Create soft link to /data/Outputs/model_logs/cvpods_playground/detection/coco/fcos/fcos.res50.fpn.coco.800size.1x 2022-03-15 15:37:57.607 | ERROR | cvpods.engine.launch:launch:64 - An error has been caught in function 'launch', process 'MainProcess' (15181), thread 'MainThread' (140337650262208): Traceback (most recent call last):

File "/icislab/volume1/wjm/cvpods/tools/train_net.py", line 165, in args=(args, config, build_model), │ │ └ <function build_model at 0x7fa13a3dd290> │ └ ╒═════════════════╤═════════════════════════════════════════════════════════════════════════════════════════════════╕ │ │ confi... └ Namespace(clearml=False, dir=None, dist_url='tcp://127.0.0.1:50206', eval_only=False, machine_rank=0, num_gpus=1, num_machine...

File "/icislab/volume1/wjm/cvpods/cvpods/engine/launch.py", line 64, in launch main_func(*args) │ └ (Namespace(clearml=False, dir=None, dist_url='tcp://127.0.0.1:50206', eval_only=False, machine_rank=0, num_gpus=1, num_machin... └ <function main at 0x7fa2e67a73b0>

File "/icislab/volume1/wjm/cvpods/tools/train_net.py", line 82, in main cfg = default_setup(config, args) │ │ └ Namespace(clearml=False, dir=None, dist_url='tcp://127.0.0.1:50206', eval_only=False, machine_rank=0, num_gpus=1, num_machine... │ └ ╒═════════════════╤═════════════════════════════════════════════════════════════════════════════════════════════════╕ │ │ confi... └ <function default_setup at 0x7fa13a3d05f0>

File "/icislab/volume1/wjm/cvpods/cvpods/engine/setup.py", line 131, in default_setup ensure_dir(output_dir) │ └ '/data/Outputs/model_logs/cvpods_playground/detection/coco/fcos/fcos.res50.fpn.coco.800size.1x' └ <function ensure_dir at 0x7fa1577545f0>

File "/icislab/volume1/wjm/cvpods/cvpods/utils/file/file_io.py", line 22, in ensure_dir megfile.smart_makedirs(path, exist_ok=True) │ │ └ '/data/Outputs/model_logs/cvpods_playground/detection/coco/fcos/fcos.res50.fpn.coco.800size.1x' │ └ <function smart_makedirs at 0x7fa157752c20> └ <module 'megfile' from '/home/weijinmeng/.local/lib/python3.7/site-packages/megfile/init.py'>

File "/home/weijinmeng/.local/lib/python3.7/site-packages/megfile/smart.py", line 443, in smart_makedirs SmartPath(path).makedirs(exist_ok) │ │ └ True │ └ '/data/Outputs/model_logs/cvpods_playground/detection/coco/fcos/fcos.res50.fpn.coco.800size.1x' └ <class 'megfile.smart_path.SmartPath'> File "/home/weijinmeng/.local/lib/python3.7/site-packages/megfile/pathlike.py", line 245, in makedirs self.mkdir(exist_ok=exist_ok) │ │ └ True │ └ <function _bind_function..smart_method at 0x7fa157cd3320> └ SmartPath('/data/Outputs/model_logs/cvpods_playground/detection/coco/fcos/fcos.res50.fpn.coco.800size.1x') File "/home/weijinmeng/.local/lib/python3.7/site-packages/megfile/smart_path.py", line 14, in smart_method return getattr(self.pathlike, name)(*args, **kwargs) │ │ │ │ └ {'exist_ok': True} │ │ │ └ () │ │ └ 'makedirs' │ └ FSPath('/data/Outputs/model_logs/cvpods_playground/detection/coco/fcos/fcos.res50.fpn.coco.800size.1x') └ SmartPath('/data/Outputs/model_logs/cvpods_playground/detection/coco/fcos/fcos.res50.fpn.coco.800size.1x') File "/home/weijinmeng/.local/lib/python3.7/site-packages/megfile/pathlike.py", line 245, in makedirs self.mkdir(exist_ok=exist_ok) │ │ └ True │ └ <function fs_makedirs at 0x7fa157c58710> └ FSPath('/data/Outputs/model_logs/cvpods_playground/detection/coco/fcos/fcos.res50.fpn.coco.800size.1x') File "/home/weijinmeng/.local/lib/python3.7/site-packages/megfile/fs_path.py", line 18, in fs_method return getattr(fs, name)(self.path_without_protocol, *args, **kwargs) │ │ │ │ │ └ {'exist_ok': True} │ │ │ │ └ () │ │ │ └ '/data/Outputs/model_logs/cvpods_playground/detection/coco/fcos/fcos.res50.fpn.coco.800size.1x' │ │ └ FSPath('/data/Outputs/model_logs/cvpods_playground/detection/coco/fcos/fcos.res50.fpn.coco.800size.1x') │ └ 'fs_makedirs' └ <module 'megfile.fs' from '/home/weijinmeng/.local/lib/python3.7/site-packages/megfile/fs.py'> File "/home/weijinmeng/.local/lib/python3.7/site-packages/megfile/fs.py", line 228, in fs_makedirs return os.makedirs(path, exist_ok=exist_ok) │ │ │ └ True │ │ └ '/data/Outputs/model_logs/cvpods_playground/detection/coco/fcos/fcos.res50.fpn.coco.800size.1x' │ └ <function makedirs at 0x7fa2e7b76ef0> └ <module 'os' from '/icislab/volume1/wjm/Anaconda/envs/open-mmlab/lib/python3.7/os.py'> File "/icislab/volume1/wjm/Anaconda/envs/open-mmlab/lib/python3.7/os.py", line 213, in makedirs makedirs(head, exist_ok=exist_ok) │ │ └ True │ └ '/data/Outputs/model_logs/cvpods_playground/detection/coco/fcos' └ <function makedirs at 0x7fa2e7b76ef0> File "/icislab/volume1/wjm/Anaconda/envs/open-mmlab/lib/python3.7/os.py", line 213, in makedirs makedirs(head, exist_ok=exist_ok) │ │ └ True │ └ '/data/Outputs/model_logs/cvpods_playground/detection/coco' └ <function makedirs at 0x7fa2e7b76ef0> File "/icislab/volume1/wjm/Anaconda/envs/open-mmlab/lib/python3.7/os.py", line 213, in makedirs makedirs(head, exist_ok=exist_ok) │ │ └ True │ └ '/data/Outputs/model_logs/cvpods_playground/detection' └ <function makedirs at 0x7fa2e7b76ef0> [Previous line repeated 4 more times] File "/icislab/volume1/wjm/Anaconda/envs/open-mmlab/lib/python3.7/os.py", line 223, in makedirs mkdir(name, mode) │ │ └ 511 │ └ '/data' └

PermissionError: [Errno 13] Permission denied: '/data'

opened by weijinmeng 2
安装cvpods出现问题
ERROR: Command errored out with exit status 1: command: /home/anaconda3/envs/OTA/bin/python -c 'import io, os, sys, setuptools, tokenize; sys.argv[0] = '"'"'/home/thomas/cvpods/setup.py'"'"'; file='"'"'/home/thomas/cvpods/setup.py'"'"';f = getattr(tokenize, '"'"'open'"'"', open)(file) if os.path.exists(file) else io.StringIO('"'"'from setuptools import setup; setup()'"'"');code = f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' develop --no-deps cwd: /home/thomas/cvpods/ Complete output (82 lines): which: no hipcc in (/home/anaconda3/envs/OTA/bin:/home/anaconda3/condabin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/root/bin) running develop running egg_info writing cvpods.egg-info/PKG-INFO writing dependency_links to cvpods.egg-info/dependency_links.txt writing requirements to cvpods.egg-info/requires.txt writing top-level names to cvpods.egg-info/top_level.txt reading manifest file 'cvpods.egg-info/SOURCES.txt' adding license file 'LICENSE' writing manifest file 'cvpods.egg-info/SOURCES.txt' running build_ext building 'cvpods._C' extension Emitting ninja build file /home/thomas/cvpods/build/temp.linux-x86_64-3.7/build.ninja... Compiling objects... Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N) [1/1] /home/anaconda3/envs/OTA/bin/x86_64-conda-linux-gnu-c++ -MMD -MF /home/thomas/cvpods/build/temp.linux-x86_64-3.7/home/thomas/cvpods/cvpods/layers/csrc/vision.o.d -Wno-unused-result -Wsign-compare -DNDEBUG -fwrapv -O2 -Wall -Wstrict-prototypes -march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-strong -fno-plt -O2 -pipe -march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-strong -fno-plt -O2 -pipe -march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-strong -fno-plt -O2 -ffunction-sections -pipe -isystem /home/anaconda3/envs/OTA/include -DNDEBUG -D_FORTIFY_SOURCE=2 -O2 -isystem /home/anaconda3/envs/OTA/include -fPIC -I/home/thomas/cvpods/cvpods/layers/csrc -I/home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/include -I/home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/include/TH -I/home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/include/THC -I/home/anaconda3/envs/OTA/include/python3.7m -c -c /home/thomas/cvpods/cvpods/layers/csrc/vision.cpp -o /home/thomas/cvpods/build/temp.linux-x86_64-3.7/home/thomas/cvpods/cvpods/layers/csrc/vision.o -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=_C -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14 FAILED: /home/thomas/cvpods/build/temp.linux-x86_64-3.7/home/thomas/cvpods/cvpods/layers/csrc/vision.o /home/anaconda3/envs/OTA/bin/x86_64-conda-linux-gnu-c++ -MMD -MF /home/thomas/cvpods/build/temp.linux-x86_64-3.7/home/thomas/cvpods/cvpods/layers/csrc/vision.o.d -Wno-unused-result -Wsign-compare -DNDEBUG -fwrapv -O2 -Wall -Wstrict-prototypes -march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-strong -fno-plt -O2 -pipe -march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-strong -fno-plt -O2 -pipe -march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-strong -fno-plt -O2 -ffunction-sections -pipe -isystem /home/anaconda3/envs/OTA/include -DNDEBUG -D_FORTIFY_SOURCE=2 -O2 -isystem /home/anaconda3/envs/OTA/include -fPIC -I/home/thomas/cvpods/cvpods/layers/csrc -I/home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/include -I/home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/include/TH -I/home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/include/THC -I/home/anaconda3/envs/OTA/include/python3.7m -c -c /home/thomas/cvpods/cvpods/layers/csrc/vision.cpp -o /home/thomas/cvpods/build/temp.linux-x86_64-3.7/home/thomas/cvpods/cvpods/layers/csrc/vision.o -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=_C -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14 cc1plus: warning: command line option '-Wstrict-prototypes' is valid for C/ObjC but not for C++ In file included from /home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/include/THC/THCGeneral.h:7, from /home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/include/THC/THC.h:4, from /home/thomas/cvpods/cvpods/layers/csrc/PSROIPool/psroi_pool_cuda.h:4, from /home/thomas/cvpods/cvpods/layers/csrc/vision.cpp:15: /home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/include/c10/cuda/CUDAStream.h:6:10: fatal error: cuda_runtime_api.h: No such file or directory 6 | #include <cuda_runtime_api.h> | ^~~~~~~~~~~~~~~~~~~~ compilation terminated. ninja: build stopped: subcommand failed. Traceback (most recent call last): File "/home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1400, in _run_ninja_build check=True) File "/home/anaconda3/envs/OTA/lib/python3.7/subprocess.py", line 512, in run output=stdout, stderr=stderr) subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.

During handling of the above exception, another exception occurred: Traceback (most recent call last): File "<string>", line 1, in <module> File "/home/thomas/cvpods/setup.py", line 192, in <module> "tools/pods_debug", File "/home/anaconda3/envs/OTA/lib/python3.7/site-packages/setuptools/__init__.py", line 153, in setup return distutils.core.setup(**attrs) File "/home/anaconda3/envs/OTA/lib/python3.7/distutils/core.py", line 148, in setup dist.run_commands() File "/home/anaconda3/envs/OTA/lib/python3.7/distutils/dist.py", line 966, in run_commands self.run_command(cmd) File "/home/anaconda3/envs/OTA/lib/python3.7/distutils/dist.py", line 985, in run_command cmd_obj.run() File "/home/anaconda3/envs/OTA/lib/python3.7/site-packages/setuptools/command/develop.py", line 34, in run self.install_for_development() File "/home/anaconda3/envs/OTA/lib/python3.7/site-packages/setuptools/command/develop.py", line 114, in install_for_development self.run_command('build_ext') File "/home/anaconda3/envs/OTA/lib/python3.7/distutils/cmd.py", line 313, in run_command self.distribution.run_command(command) File "/home/anaconda3/envs/OTA/lib/python3.7/distutils/dist.py", line 985, in run_command cmd_obj.run() File "/home/anaconda3/envs/OTA/lib/python3.7/site-packages/setuptools/command/build_ext.py", line 79, in run _build_ext.run(self) File "/root/.local/lib/python3.7/site-packages/Cython/Distutils/old_build_ext.py", line 186, in run _build_ext.build_ext.run(self) File "/home/anaconda3/envs/OTA/lib/python3.7/distutils/command/build_ext.py", line 340, in run self.build_extensions() File "/home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 580, in build_extensions build_ext.build_extensions(self) File "/root/.local/lib/python3.7/site-packages/Cython/Distutils/old_build_ext.py", line 195, in build_extensions _build_ext.build_ext.build_extensions(self) File "/home/anaconda3/envs/OTA/lib/python3.7/distutils/command/build_ext.py", line 449, in build_extensions self._build_extensions_serial() File "/home/anaconda3/envs/OTA/lib/python3.7/distutils/command/build_ext.py", line 474, in _build_extensions_serial self.build_extension(ext) File "/home/anaconda3/envs/OTA/lib/python3.7/site-packages/setuptools/command/build_ext.py", line 202, in build_extension _build_ext.build_extension(self, ext) File "/home/anaconda3/envs/OTA/lib/python3.7/distutils/command/build_ext.py", line 534, in build_extension depends=ext.depends) File "/home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 423, in unix_wrap_ninja_compile with_cuda=with_cuda) File "/home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1140, in _write_ninja_file_and_compile_objects error_prefix='Error compiling objects for extension') File "/home/anaconda3/envs/OTA/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1413, in _run_ninja_build raise RuntimeError(message) RuntimeError: Error compiling objects for extension ----------------------------------------

ERROR: Command errored out with exit status 1: /home/anaconda3/envs/OTA/bin/python -c 'import io, os, sys, setuptools, tokenize; sys.argv[0] = '"'"'/home/thomas/cvpods/setup.py'"'"'; file='"'"'/home/thomas/cvpods/setup.py'"'"';f = getattr(tokenize, '"'"'open'"'"', open)(file) if os.path.exists(file) else io.StringIO('"'"'from setuptools import setup; setup()'"'"');code = f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' develop --no-deps Check the logs for full command output. @zengarden 麻烦问下这是什么原因啊，折磨了好久了
opened by andrew-begain 0
Run on single gpu.
Good code! I am revising the code for sing gpu. and ask the of 3 lines of code in losses function(fcos.py).

if self.norm_sync: dist.all_reduce(num_foreground) num_foreground /= dist.get_world_size()

What's meaning of the 3 lines of code? Can I delete these? Thanks.
opened by kuazhangxiaoai 0

Official implementation of our CVPR2021 paper "OTA: Optimal Transport Assignment for Object Detection" in Pytorch.

Related tags

Overview

OTA: Optimal Transport Assignment for Object Detection

Requirements

Get Started

Results on COCO val set

Results on COCO test-dev

Acknowledgement

License

Comments

Reproduction

Training time for dynamic k estimation vs fixed k

how to train with one GPU?

Reason for using log scale sinkhorn algorithm

Training Error

How to solve the problem:PermissionError: [Errno 13] Permission denied: '/data'

安装cvpods出现问题

Run on single gpu.

Owner

PyTorch code for our paper "Image Super-Resolution with Non-Local Sparse Attention" (CVPR2021).

A weakly-supervised scene graph generation codebase. The implementation of our CVPR2021 paper ``Linguistic Structures as Weak Supervision for Visual Scene Graph Generation''

[PyTorch] Official implementation of CVPR2021 paper "PointDSC: Robust Point Cloud Registration using Deep Spatial Consistency". https://arxiv.org/abs/2103.05465

Code for our CVPR2021 paper coordinate attention

[CVPR2021] The source code for our paper 《Removing the Background by Adding the Background: Towards Background Robust Self-supervised Video Representation Learning》.

The official implementation of the CVPR2021 paper: Decoupled Dynamic Filter Networks

Pytorch implementation of CVPR2021 paper "MUST-GAN: Multi-level Statistics Transfer for Self-driven Person Image Generation"

A pytorch implementation of the CVPR2021 paper "VSPW: A Large-scale Dataset for Video Scene Parsing in the Wild"

offical implement of our Lifelong Person Re-Identification via Adaptive Knowledge Accumulation in CVPR2021

The official repo of the CVPR2021 oral paper: Representative Batch Normalization with Feature Calibration

Official code of paper "PGT: A Progressive Method for Training Models on Long Videos" on CVPR2021

The official PyTorch code for 'DER: Dynamically Expandable Representation for Class Incremental Learning' accepted by CVPR2021

Official implementation of Rethinking Graph Neural Architecture Search from Message-passing (CVPR2021)

Official implementation of "Towards Good Practices for Efficiently Annotating Large-Scale Image Classification Datasets" (CVPR2021)

Official implementation of our paper "LLA: Loss-aware Label Assignment for Dense Pedestrian Detection" in Pytorch.

The official pytorch implementation of our paper "Is Space-Time Attention All You Need for Video Understanding?"

The repository offers the official implementation of our paper in PyTorch.

This is the official pytorch implementation for our ICCV 2021 paper "TRAR: Routing the Attention Spans in Transformers for Visual Question Answering" on VQA Task

PyTorch code for the paper "Curriculum Graph Co-Teaching for Multi-target Domain Adaptation" (CVPR2021)