researchmm / trackit Goto Github PK

Hi! From the google drive link, I can only find the VOT2018 and VOT2019 test json. If I want to, for example, test the tracker in OTB100, where can I find the corresponding json file？

In tracking/test_ocean.py, have such a code: from tracker.ocean import Ocean
tracker in lib, while lib and tracking are sibling directories.
why can import it?
Thank you!

Thanks for your great work!
I want to know how to test with online model like OceanV18on.pth?
I have tried python tracking/test_ocean.py --arch Ocean --resume snapshot/OceanV18on.pth --dataset VOT2018 --online True but didn't work.

请问之前pysot下载的那些det、ytb、coco、vid能否直接拿来用，我看他们也是crop511的

Hi. Recently I'm experimenting SiamDW on TracKit, but could not obtain the wanted EAO as 0.280 on VOT2018.
So I open this issue for some help. Some important configurations are listed as follows.

My CUDA version is 10.2 (NVIDIA driver 440.36), and the conda env is constructed directly with your provided script in TracKit framework.

The pretrained model is collected from https://drive.google.com/file/d/1sfmy1nImNw_mstGLEOgpmJydQvSuaobI/view,
with its md5sum as 5415686007b313685065cf8f3b916742.

The hyper-parameters we utilize are collected from https://github.com/researchmm/SiamDW.

VOT2018:
    scale_step: 1.0085
    scale_lr: 0.2294
    scale_penalty: 0.9686
    w_influence: 0.1817

The raw tracking result we obtain on VOT2018 with pysot-toolkit is listed as follows, 0.013 less than the reported 0.280.

-----------------------------------------------------------------------------
|Tracker Name| Accuracy | Robustness | Lost Number |  EAO  |
-----------------------------------------------------------------------------
|   SiamDW   |  0.504   |    0.440   |    94.0     | 0.267 |
-----------------------------------------------------------------------------

I really appreciate your great work in this repo but I'd like to figure out which mistake I have made above.
Thanks in advance.

我使用提供的MSS模型在vot2020上测试eao只有0.430，我用的测试工具是0.4.2的版本，pytorch1.3，是环境问题造成的吗

Hi, Zhipeng
When I test the tracker or train it, the program always stuck or jumped out. Could you pls help me with it?
when python tracking/test_ocean.py --arch Ocean --resume snapshot/OceanV.pth --dataset VOT2019:

It just jumps out and does nothing.

when python tracking/onekey.py or python ./tracking/train_ocean.py --cfg experiments/train/Ocean.yaml --gpus 0 --workers 32 2>&1 | tee logs/Ocean_train.log:

It stuck and does not work.

the GPU status for a long time:

hi zhipeng,

  从你上篇工作SiamDW就开始关注你了，你的视频讲解也非常棒，看到你新工作赶紧来follow了~哈哈。

  切入正题，我看你数据集最终形式是xxx_crop511的，后续会考虑更新开源相关代码或工具，让我们直接从本地已有各个数据集（如GOT10K）crop出图像对吗（也就是最终生成你提供的xxx_crop511形式数据集）？

  感谢回复！

Hi,
When I try to download the pre-trained PyTorch model from google drive, it told me 'download quota exceeded for this file so you can not download it at this time ', are there any other resources I can download?

Hi Zhipeng

It seems that I can not setup the conda env successfully, help me pls. Thanks!

您好，我用一张GPU尝试tune_tpe，跑了几个小时候显示：RuntimeError: CUDA out of memory. Tried to allocate 90.00 MiB (GPU 0; 10.76 GiB total capacity; 6.31 GiB already allocated; 22.12 MiB free; 6.36 GiB reserved in total by PyTorch)
请问，对于tune tpe一张显卡是否不够用？

如题

hi zhipeng,

Sorry to bother you again, I guess this question will confuse some people like me who have to run your code on Windows, so I'm using English and hope more people can join the discussion.

problem description

Due to some reasons, i have to run your code(Ocean) on win10(two 2080ti gpus) , when i had prepared the datasets and corresponding env.

After run the code , i encounter the following mistakes:

  Traceback (most recent call last):
  File "D:/hwh/code/TracKit/tracking/train_ocean.py", line 21, in <module>
    import models.models as models
  File "D:\hwh\code\TracKit\tracking\..\lib\models\models.py", line 14, in <module>
    from .connect import box_tower, AdjustLayer, AlignHead, Corr_Up, MultiDiCorr, OceanCorr
  File "D:\hwh\code\TracKit\tracking\..\lib\models\connect.py", line 9, in <module>
    from .dcn import DeformConv, DeformConvPack
  File "D:\hwh\code\TracKit\tracking\..\lib\models\dcn\__init__.py", line 1, in <module>
    from .deform_conv import (DeformConv, DeformConvPack, ModulatedDeformConv,
  File "D:\hwh\code\TracKit\tracking\..\lib\models\dcn\deform_conv.py", line 9, in <module>
    from . import deform_conv_cuda
**ImportError: cannot import name 'deform_conv_cuda' from 'models.dcn'** (D:\hwh\code\TracKit\tracking\..\lib\models\dcn\__init__.py)

for help

i noticed that this problems is due to the file deform_conv_cuda.cpython-37m-x86_64-linux-gnu.so is only can be import on the os for linux .

so, could you help me to solve this problem? or tell me how to import this moudle on windows?

thanks for reply! (i have to eat something now , this problem caused me to miss lunch ,hah~)

您好，我现在做一些改进分别在VOT2018和GOT-10k上测试，然后有些改进对其中一个数据集有增长但是另一个数据集会掉点，请问应该以哪个为标准，还有好像在GOT-10k的val和test上好像差异也挺大的，我试了一些方法在val上涨点但是在test上掉点，反过来的情况也出现过

Have you ever consider use a Alexnet backbone? Because I found "alex" in the config yaml file.

首先感谢您的工作，请问这个可以完成多目标的任务吗，对多目标进行分割和跟踪

Hi, since the size of training data is large (y2b is up to 130+GB), it is quite hard to directly download in the browser. I also try to use command-line method. (https://www.toodyao.com/?p=3846) But I can't obtain FILEID. (when I click y2b, the returned google drive link is shown as below, https://drive.google.com/file/d/1XntleNeFDKNS9kdspn9xY7UWuV2USCmt/view?usp=sharing. I can't find FILEID).
So can you help me? or are there alternative methods to download these data? Thank you

Thank you for your excellent work. I have a problem about testing on LaSOT. On the paper of LaSOT: A High-quality Benchmark for Large-scale Single Object Tracking, it has proposed two evaluation protocols. Do you follow the second evalution protocol: train the tracker only using LaSOT training subset and test on the testing subset?

I want to know diff between eval and test.
If I run a demo, is there existing code? Or which file I should change?
Thank you!

there is only how to install tensorrt, but would you please help to provide the code with tensorrt demo? thanks.

您好，很棒的工作，但我在下载完您提供的y2b.tar训练数据后，发现train.json里头的bbox标注都是小于等于 1.0的 ， 这是一种中间格式吗 还是说有些问题？

需要改哪里

我对最速下降中的 distancemap不是很理解，原理和计算步骤不太懂。
请问，它的物理意义是什么？在代码中它输出torch.size([30,100,14,14])的tensor，
30维度表示图像数量，后面三个维度就不懂了。
希望您解答下，谢谢！

Hi, thanks for your work.
I got the "DeprecationWarning: The redirect_output argument is deprecated." when I tune the training models.
It seems that the latest version of ray is not suitable.
Could you show us which version of ray do you use?
And is there any more requirements for tuning?

作者您好，当我在评估vot测验结果的时候，报错如下：
File "/home/ubuntu/TracKit/lib/eval_toolkit/bin/../../eval_toolkit/pysot/evaluation/eao_benchmark.py", line 110, in _calculate_eao
max_len = max([len(x) for x in all_overlaps])
ValueError: max() arg is an empty sequence
请问可以解答一下什么原因吗？谢谢

I can not see your paper in ECCV2020 accepted paper list?

Hi

I trained the model from scratch using the instructions you provided (link). I can see the training setting in the yaml file (link) is the same as mentioned in the paper. The training script seems to work fine, but when I test the trained model, I don't get the same results as reported in the paper.

I run the training using following command:

python tracking/onekey.py

The training curve I obtain is as folllowing:

I run the testing using the checkpoint form the last epoch:

python tracking/test_ocean.py --arch Ocean --resume snapshot/checkpoint_e50.pth --dataset VOT2019

I run the evaluation on the obtained testing results:

python lib/eval_toolkit/bin/eval.py --dataset_dir dataset --dataset VOT2019 --tracker_result_dir result/VOT2019 --trackers Ocean

The results I obtain are as following:

As you can see they differ quite a lot from the results reported in the paper (Accuracy: 0.590, Robustness: 0.376, EAO: 0.327), and also from the results I obtain using your provided checkpoint. Is there any explanation for this ?

请问这个问题怎么解决，我没有看到torch2trt文件呀

Is pretrain.model trained on standard ResNet50? What is the difference with this version https://github.com/ohhhyeahhh/SiamCAR#download-pretrained-backbones?
thanks!

hi zhipeng：
在编译期间遇到了一个问题请问这个应该怎么解决呢，如下：
(pysot) xxxi@ubuntu:~/Code/TracKit$ python setup.py develop
/home/xxxi/anaconda3/envs/pysot/lib/python3.7/site-packages/setuptools/dist.py:452: UserWarning: Normalizing '1.0.rc0' to '1.0rc0'
warnings.warn(tmpl.format(**locals()))
running develop
running egg_info
writing mmdet.egg-info/PKG-INFO
writing dependency_links to mmdet.egg-info/dependency_links.txt
writing top-level names to mmdet.egg-info/top_level.txt
reading manifest file 'mmdet.egg-info/SOURCES.txt'
writing manifest file 'mmdet.egg-info/SOURCES.txt'
running build_ext
building 'lib.models.dcn.deform_conv_cuda' extension
Emitting ninja build file /home/zhuyi/Code/TracKit/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)
[1/2] /usr/local/cuda/bin/nvcc -I/home/zhuyi/anaconda3/envs/pysot/lib/python3.7/site-packages/torch/include -I/home/xxxi/anaconda3/envs/pysot/lib/python3.7/site-packag es/torch/include/torch/csrc/api/include -I/home/zhuyi/anaconda3/envs/pysot/lib/python3.7/site-packages/torch/include/TH -I/home/xxxi/anaconda3/envs/pysot/lib/python3.7 /site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/zhuyi/anaconda3/envs/pysot/include/python3.7m -c -c /home/xxxi/Code/TracKit/lib/models/dcn/src/defor m_conv_cuda_kernel.cu -o /home/xxxi/Code/TracKit/build/temp.linux-x86_64-3.7/lib/models/dcn/src/deform_conv_cuda_kernel.o -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF CONVERSIONS_ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr --compiler-options ''"'"'-fPIC'"'"'' -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ - D__CUDA_NO_HALF2_OPERATORS__ -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=deform_conv_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_86,code=sm_86 -std =c++14
FAILED: /home/xxxi/Code/TracKit/build/temp.linux-x86_64-3.7/lib/models/dcn/src/deform_conv_cuda_kernel.o
/usr/local/cuda/bin/nvcc -I/home/zhuyi/anaconda3/envs/pysot/lib/python3.7/site-packages/torch/include -I/home/xxxi/anaconda3/envs/pysot/lib/python3.7/site-packages/tor ch/include/torch/csrc/api/include -I/home/zhuyi/anaconda3/envs/pysot/lib/python3.7/site-packages/torch/include/TH -I/home/xxxi/anaconda3/envs/pysot/lib/python3.7/site- packages/torch/include/THC -I/usr/local/cuda/include -I/home/zhuyi/anaconda3/envs/pysot/include/python3.7m -c -c /home/xxxi/Code/TracKit/lib/models/dcn/src/deform_conv cuda_kernel.cu -o /home/xxxi/Code/TracKit/build/temp.linux-x86_64-3.7/lib/models/dcn/src/deform_conv_cuda_kernel.o -D__CUDA_NO_HALF_OPERATORS_ -D__CUDA_NO_HALF_CONVE RSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr --compiler-options ''"'"'-fPIC'"'"'' -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUD A_NO_HALF2_OPERATORS__ -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=deform_conv_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_86,code=sm_86 -std=c++14
nvcc fatal : Unsupported gpu architecture 'compute_86'
我的配置是RTX3090 cuda11 cudnn8.0

I'm not understand what grid in Ocean_ does. Do not understand how score_size in Ocean_ is calculated?
For the template of 127, featmap size becomes 15 after feature extract;
for search image of 255, featmap size becomes 31 after feature extract.
Why the stride of backbone is 8?
Can you tell me a little bit about grid?
And how is score_size calculated in detail?
Thank you!

请问，第一行与第二行数据是一样的吗？是都需要下载还是只需要下载一行就好了?
之后，下载完数据后，要跑哪些脚本呢？
谢谢！

hi zhipeng,请问是否有打算将训练数据集上传百度云盘

I am doing some work base on the no object-aware verison.
I wonder what‘s the results of OceanNoOA.pth in VOT2018 and VOT2019?
What I get from the paper is that the EAO should be 0.438 in VOT2018. But in my test, it is much lower.
And I didn't find the VOT2019 results in the paper for OceanNoOA.pth.
Thanks for your help!

您好, 向您请教一个小问题. 为什么训练模型时不做验证, 这与机器学习的一般做法很不一样? 不会担心过拟合吗? 我看到您在30个epoch之后开始 test, 有没有可能在第30个 epoch 之前就出现过拟合呢?
我发现其它 Siamese tracker 的代码中也存在这样的情况, 作为新人我不太理解这种做法.

Hi, thanks for your excellent work. I notice that Ocean trained with Youtube-BB, VID, DET, GOT-10K and COCO on 8 Tesla V100 GPUs. Not everyone could afford the perfect hardware resources and huge datasets. I wonder how long time does it take on training?

I am setting up the environment as mentioned in the documentation. I was able to successfully run the following steps

cd $TracKit/lib/tutorial
bash install.sh $conda_path TracKit
cd $TracKit
conda activate TracKit

But failing on
python setup.py develop

Environment details
Python 3.7.9
[GCC 7.3.0] :: Anaconda, Inc. on linux
Cuda: 10.2
Graphics card: GeForce GTX TITAN X
OS: Ubuntu 18.04 LTS

The complete error is as follows.

(TracKit) deepen@deepen-Z11PA-U12-Series:~/SandeepMenon/TracKit$ python setup.py develop
/home/deepen/anaconda3/envs/TracKit/lib/python3.7/site-packages/setuptools/dist.py:452: UserWarning: Normalizing '1.0.rc0' to '1.0rc0'
  warnings.warn(tmpl.format(**locals()))
running develop
running egg_info
writing mmdet.egg-info/PKG-INFO
writing dependency_links to mmdet.egg-info/dependency_links.txt
writing top-level names to mmdet.egg-info/top_level.txt
reading manifest file 'mmdet.egg-info/SOURCES.txt'
writing manifest file 'mmdet.egg-info/SOURCES.txt'
running build_ext
building 'lib.models.dcn.deform_conv_cuda' extension
Emitting ninja build file /home/deepen/SandeepMenon/TracKit/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] c++ -MMD -MF /home/deepen/SandeepMenon/TracKit/build/temp.linux-x86_64-3.7/lib/models/dcn/src/deform_conv_cuda.o.d -pthread -B /home/deepen/anaconda3/envs/TracKit/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/home/deepen/.local/lib/python3.7/site-packages/torch/include -I/home/deepen/.local/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/deepen/.local/lib/python3.7/site-packages/torch/include/TH -I/home/deepen/.local/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/deepen/anaconda3/envs/TracKit/include/python3.7m -c -c /home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp -o /home/deepen/SandeepMenon/TracKit/build/temp.linux-x86_64-3.7/lib/models/dcn/src/deform_conv_cuda.o -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=deform_conv_cuda -D_GLIBCXX_USE_CXX11_ABI=0 -std=c++14
FAILED: /home/deepen/SandeepMenon/TracKit/build/temp.linux-x86_64-3.7/lib/models/dcn/src/deform_conv_cuda.o 
c++ -MMD -MF /home/deepen/SandeepMenon/TracKit/build/temp.linux-x86_64-3.7/lib/models/dcn/src/deform_conv_cuda.o.d -pthread -B /home/deepen/anaconda3/envs/TracKit/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -I/home/deepen/.local/lib/python3.7/site-packages/torch/include -I/home/deepen/.local/lib/python3.7/site-packages/torch/include/torch/csrc/api/include -I/home/deepen/.local/lib/python3.7/site-packages/torch/include/TH -I/home/deepen/.local/lib/python3.7/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/deepen/anaconda3/envs/TracKit/include/python3.7m -c -c /home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp -o /home/deepen/SandeepMenon/TracKit/build/temp.linux-x86_64-3.7/lib/models/dcn/src/deform_conv_cuda.o -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=deform_conv_cuda -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++
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp: In function ‘void shape_check(at::Tensor, at::Tensor, at::Tensor*, at::Tensor, int, int, int, int, int, int, int, int, int, int)’:
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp:63:3: error: ‘AT_CHECK’ was not declared in this scope
   AT_CHECK(weight.ndimension() == 4,
   ^~~~~~~~
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp:63:3: note: suggested alternative: ‘DCHECK’
   AT_CHECK(weight.ndimension() == 4,
   ^~~~~~~~
   DCHECK
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp: In function ‘int deform_conv_forward_cuda(at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, int, int, int, int, int, int, int, int, int, int, int)’:
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp:190:3: error: ‘AT_CHECK’ was not declared in this scope
   AT_CHECK((offset.size(0) == batchSize), "invalid batch size of offset");
   ^~~~~~~~
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp:190:3: note: suggested alternative: ‘DCHECK’
   AT_CHECK((offset.size(0) == batchSize), "invalid batch size of offset");
   ^~~~~~~~
   DCHECK
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp: In function ‘int deform_conv_backward_input_cuda(at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, int, int, int, int, int, int, int, int, int, int, int)’:
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp:296:3: error: ‘AT_CHECK’ was not declared in this scope
   AT_CHECK((offset.size(0) == batchSize), 3, "invalid batch size of offset");
   ^~~~~~~~
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp:296:3: note: suggested alternative: ‘DCHECK’
   AT_CHECK((offset.size(0) == batchSize), 3, "invalid batch size of offset");
   ^~~~~~~~
   DCHECK
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp: In function ‘int deform_conv_backward_parameters_cuda(at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, int, int, int, int, int, int, int, int, int, int, float, int)’:
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp:411:3: error: ‘AT_CHECK’ was not declared in this scope
   AT_CHECK((offset.size(0) == batchSize), "invalid batch size of offset");
   ^~~~~~~~
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp:411:3: note: suggested alternative: ‘DCHECK’
   AT_CHECK((offset.size(0) == batchSize), "invalid batch size of offset");
   ^~~~~~~~
   DCHECK
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp: In function ‘void modulated_deform_conv_cuda_forward(at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, int, int, int, int, int, int, int, int, int, int, bool)’:
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp:491:3: error: ‘AT_CHECK’ was not declared in this scope
   AT_CHECK(input.is_contiguous(), "input tensor has to be contiguous");
   ^~~~~~~~
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp:491:3: note: suggested alternative: ‘DCHECK’
   AT_CHECK(input.is_contiguous(), "input tensor has to be contiguous");
   ^~~~~~~~
   DCHECK
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp: In function ‘void modulated_deform_conv_cuda_backward(at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, at::Tensor, int, int, int, int, int, int, int, int, int, int, bool)’:
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp:572:3: error: ‘AT_CHECK’ was not declared in this scope
   AT_CHECK(input.is_contiguous(), "input tensor has to be contiguous");
   ^~~~~~~~
/home/deepen/SandeepMenon/TracKit/lib/models/dcn/src/deform_conv_cuda.cpp:572:3: note: suggested alternative: ‘DCHECK’
   AT_CHECK(input.is_contiguous(), "input tensor has to be contiguous");
   ^~~~~~~~
   DCHECK
ninja: build stopped: subcommand failed.
Traceback (most recent call last):
  File "/home/deepen/.local/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1400, in _run_ninja_build
    check=True)
  File "/home/deepen/anaconda3/envs/TracKit/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 "setup.py", line 174, in <module>
    zip_safe=False)
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/site-packages/setuptools/__init__.py", line 153, in setup
    return distutils.core.setup(**attrs)
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/distutils/core.py", line 148, in setup
    dist.run_commands()
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/distutils/dist.py", line 966, in run_commands
    self.run_command(cmd)
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/distutils/dist.py", line 985, in run_command
    cmd_obj.run()
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/site-packages/setuptools/command/develop.py", line 34, in run
    self.install_for_development()
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/site-packages/setuptools/command/develop.py", line 136, in install_for_development
    self.run_command('build_ext')
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/distutils/cmd.py", line 313, in run_command
    self.distribution.run_command(command)
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/distutils/dist.py", line 985, in run_command
    cmd_obj.run()
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/site-packages/setuptools/command/build_ext.py", line 79, in run
    _build_ext.run(self)
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/site-packages/Cython/Distutils/old_build_ext.py", line 186, in run
    _build_ext.build_ext.run(self)
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/distutils/command/build_ext.py", line 340, in run
    self.build_extensions()
  File "/home/deepen/.local/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 580, in build_extensions
    build_ext.build_extensions(self)
  File "/home/deepen/anaconda3/envs/TracKit/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/deepen/anaconda3/envs/TracKit/lib/python3.7/distutils/command/build_ext.py", line 449, in build_extensions
    self._build_extensions_serial()
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/distutils/command/build_ext.py", line 474, in _build_extensions_serial
    self.build_extension(ext)
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/site-packages/setuptools/command/build_ext.py", line 196, in build_extension
    _build_ext.build_extension(self, ext)
  File "/home/deepen/anaconda3/envs/TracKit/lib/python3.7/distutils/command/build_ext.py", line 534, in build_extension
    depends=ext.depends)
  File "/home/deepen/.local/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 423, in unix_wrap_ninja_compile
    with_cuda=with_cuda)
  File "/home/deepen/.local/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/deepen/.local/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

Am I missing something? Any help is appreciated.
Thank you

Hello,my environment is 4 1080ti, every time I run, if I use all the data sets, it will be killed. Is there any good way to solve this problem?

Hi,
Thanks for your excellent work.
When will you release the ocean plus code and paper?

Best

问的问题比较幼稚，但是实在是对您的工作很感兴趣

PermissionError: [Errno 13] Permission denied: '/tmp/torch_extensions/_prroi_pooling'
在使用online时，报上述错误，因为PrRoIPooling是用c写的，我没有找到如何更改路径到用户目录下。
您有解决方法吗？
还有，我在阅读该源码时，绕来绕去的，非常乱，您有什么好的建议吗？更进一步，您在写出这么优秀的项目，经过了哪些学习呢？有推荐的课程/书籍/方法吗？
谢谢！

请问你们这个网络要训练多久？

Save PyTorch model to baidu-disk link? Using the Google cloud is not friendly.
Thank you!

Hi,

I can see that you have provided the json files needed to do evaluation on VOT2019. But there are no such files for got-10k dataset. Can you please provide those as well ?

Thanks

Hi, thanks for your work, I have some problem in training.
I find that in TracKit/experiments/train/Ocean.yaml this model train for 50 epoch, but in the paper training for 20 epoch, and the learing rate end_lr also seems for alexnet training.
Another problem is that I only have one 2080ti gpu, will it influence the results a lot? Or should I adjust some parameters.
Thanks.

Hi,

In the tutorial for Ocean, you mentioned that the testing can be done on LASOT as well. But you don't provide the corresponding yaml file (/experiments/test/LASOT/Ocean.yaml).

Can you share this yaml file, please ?

I encountered an error when I ran the test_siamdw file:
python tracking/test_siamdw.py --arch Ocean --resume snapshot/siamdw_res22w.pth --dataset VOT2019
Traceback (most recent call last):
File "tracking/test_siamdw.py", line 13, in
import models.models as models
File "/home/xiaobai/PycharmProjects/ObjectTrackingCode/DLCode/TracKit-master/tracking/../lib/models/models.py", line 14, in
from .connect import box_tower, AdjustLayer, AlignHead, Corr_Up, MultiDiCorr, OceanCorr
File "/home/xiaobai/PycharmProjects/ObjectTrackingCode/DLCode/TracKit-master/tracking/../lib/models/connect.py", line 9, in
from .dcn import DeformConv, DeformConvPack
File "/home/xiaobai/PycharmProjects/ObjectTrackingCode/DLCode/TracKit-master/tracking/../lib/models/dcn/init.py", line 1, in
from .deform_conv import (DeformConv, DeformConvPack, ModulatedDeformConv,
File "/home/xiaobai/PycharmProjects/ObjectTrackingCode/DLCode/TracKit-master/tracking/../lib/models/dcn/deform_conv.py", line 9, in
from . import deform_conv_cuda
ImportError: libcudart.so.9.0: cannot open shared object file: No such file or directory.

Could you tall me how can i solve it? Thanks a lot.

If I use "python lib/eval_toolkit/bin/eval.py " to eval the performance of model, which datasets are supported? Only VOT or some of VOT, OTB, GOT10K, LASOT ? Thanks !

Thanks for sharing!
Wanting to test UAV123 dataset, could you pls provide the UAV.json file?
Or just the toolkit that can generate the json file from txt files. Thanks advanced!

我发现论文中的在线更新模块需要train一个与resnet相同的block，以及DiMP需要的初始化参数。但我在代码中没能找到如何训练该分支的方法，能否告知我如何训练该分支，
这个分支是和离线的孪生网络共同训练的吗？还是先训练离线网络，再训练在线网络？
谢谢。