![Which type of machine are you using? No distributed training Do you want to run your training on CPU only (even if a GPU is available)? [yes/NO]: Do you wish to optimize your script with torch dynamo? [yes/NO]: Do you want to use DeepSpeed? [yes/NO]: What GPU(s) (by id) should be used for training on this machine as a comma-seperated list?. Funky](/img/512x512/1544061104138.webp)
dist_util.setup_dist()---> RuntimeError: Distributed package doesn't have NCCL built in 👍 3 nathanterroir, kbatsuren, and TneitaP reacted with thumbs up emoji All reactionsHost and manage packages Security. Find and fix vulnerabilities Codespaces. Instant dev environments ... RuntimeError: Distributed package doesn't have NCCL built inRuntimeError: Distributed package doesn't have NCCL built in. distributed. 27: 9691: August 30, 2023 RuntimeError: CUDA out of memory. Tried to allocate - Can I solve ...May 11, 2022 · Distributed package doesn't have NCCL built in. 问题描述： python在windows环境下dist.init_process_group(backend, rank, world_size)处报错‘RuntimeError: Distributed package doesn’t have NCCL built in’，具体信息如下： Mar 22, 2023 · 这篇文章可能适合什么读者：对sovits的复现感兴趣，但本地设备显卡算力不足，打算通过autodl等平台租借显卡，在anaconda+linuxs平台上复现sovits4.0的读者。. （虽然后文也有涉及一点win系统上复现可能出现问题）. 以下内容视作读者具备基本的代码复现知识，不过 ... RuntimeError: Distributed package doesn't have NCCL built in #5. RuntimeError: Distributed package doesn't have NCCL built in. #5. Closed. AIisCool opened this issue on Aug 19, 2022 · 1 comment. qiuzhongwei-USTB closed this as completed on Dec 13, 2022. Sign up for free to join this conversation on GitHub .To rebuild or reinstall the package, you can follow the directions in the documentation of the relevant framework. Verify GPU drivers: Ensure your computer has the necessary GPU drivers installed. For NCCL to work appropriately, suitable GPU drivers are needed.PyTorchのCUDAプログラミングに絞って並列処理を見てみる。. なお、 CPU側の並列処理は別資料に記載済みである 。. ここでは、. C++の拡張仕様であるCUDAの基礎知識. カーネルレベルの並列処理. add関数の実装. im2col関数の実装. ストリームレベルの並列処理 ... Mar 8, 2021 · dist_util.setup_dist()---> RuntimeError: Distributed package doesn't have NCCL built in 👍 3 nathanterroir, kbatsuren, and TneitaP reacted with thumbs up emoji All reactions System Info PyTorch version : 2.01 and nightly NVIDIA-SMI 535.54.03 Driver Version: 535.54.03 CUDA Version: 12.2 I installed cuda 11.8 with conda by pip install -r requirements.txt . Ubuntu 2204 wi...edited. Install CUDA's latest toolkit 10.1 and equivalent CuDNN 7.5.1. Install Openmpi v3.1.2 with CUDA support. Build / install pytroch from source. Test any communication for a process group with mpi backend. PyTorch Version (e.g., 1.0): 1.1. OS (e.g., Linux): Ubuntu 16.04. How you installed PyTorch ( conda, pip, source): installed from ...Aug 9, 2021 · How to train a custom model under Windows 10 with miniconda? Inference works great but when I try to start a custom training only errors come up. Latest RTX/Quadro driver and Nvida Cuda Toolkit 11.3 + cudnn 11.3 + ms vs buildtools are in... {"payload":{"allShortcutsEnabled":false,"fileTree":{"torch/distributed":{"items":[{"name":"_composable","path":"torch/distributed/_composable","contentType ...failure to initialize NCCL. #216. Open. metaphorz opened this issue on Mar 18, 2021 · 3 comments.raise RuntimeError("Distributed package doesn’t have NCCL "RuntimeError: Distributed package doesn’t have NCCL built in. All these errors are raised when the init_process_group() function is called as following: torch.distributed.init_process_group(backend='nccl', init_method=args.dist_url, world_size=args.world_size, rank=args.rank)Aug 14, 2023 · raise RuntimeError("Distributed package doesn't have NCCL " "built in") RuntimeError: Distributed package doesn't have NCCL built in During handling of the above exception, another exception occurred: The multiprocessing and distributed confusing me a lot when I’m reading some code. #the main function to enter def main_worker (rank,cfg): trainer=Train (rank,cfg) if __name__=='_main__': torch.mp.spawn (main_worker,nprocs=cfg.gpus,args= (cfg,)) #here is a slice of Train class class Train (): def __init__ (self,rank,cfg): #nothing special if ...raise RuntimeError(“Distributed package doesn‘t have NCCL “ “built in“) RuntimeError: Distributed pa_lanmy_dl的博客-程序员秘密. 技术标签： 训练过程 安装配置 python ubuntu pytorch 服务器When trying to run example_completion.py file in my windows laptop, I am getting below error: I am using pytorch 2.0 version with CUDA 11.7 . On typing the command import torch.distributed as dist ...Nov 2, 2018 · RuntimeError: Distributed package doesn’t have NCCL built in I install pytorch from the source v1.0rc1, getting the config summary as follows: USE_NCCL is On, Private Dependencies does not include nccl, nccl is not built-in. Aug 24, 2021 · Start multiple jobs on one computer. You need to specify a different port for each job (29500 by default) to avoid communication conflict. the solution is to specify the port while running the program, and give the port number arbitrarily before the PY file to be executed: python -m torch.distributed.launch --nproc_per_node=1 --master_port ... {"payload":{"allShortcutsEnabled":false,"fileTree":{"torch/distributed":{"items":[{"name":"_composable","path":"torch/distributed/_composable","contentType ...Here is a good NVIDIA installation guide. Check if you already have an NVIDIA driver with nvidia-smi.. If you already have the NVIDIA drivers correctly installed, install PyTorch from the official source according to your system.372 raise RuntimeError("Distributed package doesn't have NCCL " 373 "built in" ) 374 _default_pg = ProcessGroupNCCL(store, rank, world_size)Runtimeerror: distributed package doesn’t have nccl built in May 12, 2023 by adones evangelista When working with distributed computing and parallel processing, encountering errors is not uncommon.RuntimeError: Distributed package doesn't have NCCL built in. The text was updated successfully, but these errors were encountered: All reactions. Copy link ...Learn more » Push, build, and install RubyGems npm packages Python packages Maven artifacts PHP packages Go Modules Bower components Debian packages RPM packages NuGet packages.错误: RuntimeError: Distributed package doesn't have NCCL built in|PyTorch踩坑. bug / PyTorch 2021-09-28 赵亚博([email protected]). Read more >RuntimeError: Distributed package doesn’t have NCCL built in All these errors are raised when the init_process_group () function is called as following: torch.distributed.init_process_group (backend='nccl', init_method=args.dist_url, world_size=args.world_size, rank=args.rank) Here, note that args.world_size=1 and rank=args.rank=0.Apr 2, 2023 · RuntimeError: The disk is in use or locked by another process. I am trying out the code for the paper "SinDiffusion". When I try to run this code as said in the read.me file, : mpiexec -n 8 python image_train.py --data_dir data/image1.png --lr 5e-4 --diffusion_steps 1000 --image_size 256 --noise_schedule linear --num_channels 64 --num_head ... You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window.Aug 19, 2022 · Hi, nngg11, I'm not sure if this codebase supports training / testing on windows since I have never tried this before. I only use linux-based systems, and I guess there will be some problems if you run training / testing on windows. Windows RuntimeError: Distributed package doesn‘t have NCCL built in问题; pytorchlighting报错：raise RuntimeError(“Distributed package doesn‘t have NCCL “RuntimeError: Distribu; Mybatis报错“Field ‘id‘ doesn‘t have a default value” 由sklearn doesn't have attribute 'datasets'引发的思考 这篇文章可能适合什么读者：对sovits的复现感兴趣，但本地设备显卡算力不足，打算通过autodl等平台租借显卡，在anaconda+linuxs平台上复现sovits4.0的读者。. （虽然后文也有涉及一点win系统上复现可能出现问题）. 以下内容视作读者具备基本的代码复现知识，不过 ...Sep 12, 2022 · Hi, thanks for taking time and mentioning these useful tips . I am very sorry for the late reply cause I was checking my computer and source code. Sep 23, 2022 · Distributed environment: MULTI_GPU Backend: nccl Num processes: 2 Process index: 1 Local process index: 1 Device: cuda:1 Distributed environment: MULTI_GPU Backend: nccl Num processes: 2 Process index: 0 Local process index: 0 Device: cuda:0 Could you please share what hardware you’re running on and what env? To rebuild or reinstall the package, you can follow the directions in the documentation of the relevant framework. Verify GPU drivers: Ensure your computer has the necessary GPU drivers installed. For NCCL to work appropriately, suitable GPU drivers are needed.pytorchlighting报错：raise RuntimeError(“Distributed package doesn‘t have NCCL “RuntimeError: Distribu，代码先锋网，一个为软件开发程序员提供代码片段和技术文章聚合的网站。 Jan 6, 2022 · Cause: use mmdetection’s tools/benchmark An error occurs when py calculates FPS the error contents are as follows: Traceback (most recent call last): File "tools ... Dec 12, 2022 · Here is a good NVIDIA installation guide. Check if you already have an NVIDIA driver with nvidia-smi.. If you already have the NVIDIA drivers correctly installed, install PyTorch from the official source according to your system. You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window.RuntimeError: Distributed package doesn't have NCCL built in #112 Open Distributed package doesn't have NCCL / The requested address is not valid in its context.Mar 22, 2023 · 这篇文章可能适合什么读者：对sovits的复现感兴趣，但本地设备显卡算力不足，打算通过autodl等平台租借显卡，在anaconda+linuxs平台上复现sovits4.0的读者。. （虽然后文也有涉及一点win系统上复现可能出现问题）. 以下内容视作读者具备基本的代码复现知识，不过 ... You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window.Jul 17, 2022 · RuntimeError: Distributed package doesn't have NCCL built in Traceback (most recent call last): File "tools/train.py", line 250, in main() raise RuntimeError("Distributed package doesn't have NCCL " "built in") RuntimeError: Distributed package doesn't have NCCL built in Any help would be greatly appreciated, and I have no problem compensating anyone who can help me solve this issue. Runtimeerror: distributed package doesn’t have nccl built in May 12, 2023 by adones evangelista When working with distributed computing and parallel processing, encountering errors is not uncommon.Aug 12, 2021 · As the accelerate command was not working from poershell, I used the torch.distributed.launch to run the script as follows: python -m torch.distributed.launch --nproc_per_node 1 --use_env ./nlp_example.py Since I was using Windows OS, it gave the following error: RuntimeError: Distributed package doesn't have NCCL built in Jun 5, 2021 · This entry was posted in How to Fix and tagged distributed package doesn't have nccl error, ProgrammerAH on 2021-06-05 by Robins. Post navigation ← Flutter Package error: keyboard_visibility:verifyReleaseResources How to Solve error: command ‘C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.0\bin vcc.exe‘ failed → Distributed package doesn’t have NCCL built in Hi @nguyenngocdat1995 , sorry for the delay - Jetson doesn’t have NCCL, as this library is intended for multi-node servers. You may need to disable the multiprocessing in the detectron’s training.Here is a good NVIDIA installation guide. Check if you already have an NVIDIA driver with nvidia-smi.. If you already have the NVIDIA drivers correctly installed, install PyTorch from the official source according to your system.The distributed package comes with a distributed key-value store, which can be used to share information between processes in the group as well as to initialize the distributed package in torch.distributed.init_process_group () (by explicitly creating the store as an alternative to specifying init_method .)Jul 17, 2022 · RuntimeError: Distributed package doesn't have NCCL built in Traceback (most recent call last): File "tools/train.py", line 250, in main() Jul 6, 2022 · python.distributedは、Point-to-Point通信や集団通信といった分散処理のAPIを提供しています。これにより、細かな処理をカスタマイズすることが可能です。 通信のbackendとしては、pytorch 1.13時点では、MPI、GLOO、NCCLが選択できます。各backendで利用できる通信関数の一覧は公式ドキュメントに記載されて ... RuntimeError: Distributed package doesn't have NCCL built in. distributed. 23: 8639: August 22, 2023 ← previous page next page ...Aug 19, 2022 · RuntimeError: Distributed package doesn't have NCCL built in #5. RuntimeError: Distributed package doesn't have NCCL built in. #5. Closed. AIisCool opened this issue on Aug 19, 2022 · 1 comment. qiuzhongwei-USTB closed this as completed on Dec 13, 2022. Sign up for free to join this conversation on GitHub . python.distributedは、Point-to-Point通信や集団通信といった分散処理のAPIを提供しています。これにより、細かな処理をカスタマイズすることが可能です。 通信のbackendとしては、pytorch 1.13時点では、MPI、GLOO、NCCLが選択できます。各backendで利用できる通信関数の一覧は公式ドキュメントに記載されて ...Runtimeerror: distributed package doesn’t have nccl built in; Runtimeerror: context has already been set; cuda error: all cuda-capable devices are busy or unavailable;Apr 1, 2022 · [Solved] mmdetection benchmark.py Error: RuntimeError: Distributed package doesn‘t have NCCL built in; RuntimeError: Address already in use [How to Solve] Brew install XXX and display error: [email protected] [How to Solve] [Solved] RuntimeError: Numpy is not available (Associated Torch or Tensorflow) RuntimeError: Distributed package doesn't have NCCL built in #6. RuntimeError: Distributed package doesn't have NCCL built in. #6. Open. juntao66 opened this issue on May 1, 2021 · 4 comments.Jan 8, 2011 · 372 raise RuntimeError("Distributed package doesn't have NCCL " 373 "built in" ) 374 _default_pg = ProcessGroupNCCL(store, rank, world_size) Apr 4, 2021 · Please add a note for "Fit More and Train Faster With ZeRO via DeepSpeed and FairScale" that deepspeed or parallel training is not easy/possible on Windows (10 for me) as nccl is not supported (directly) on windows yet.. Which type of machine are you using? No distributed training Do you want to run your training on CPU only (even if a GPU is available)? [yes/NO]: Do you wish to optimize your script with torch dynamo? [yes/NO]: Do you want to use DeepSpeed? [yes/NO]: What GPU(s) (by id) should be used for training on this machine as a comma-seperated list?I tried printing the issue with os.environ["TORCH_DISTRIBUTED_DEBUG"]="DETAIL" it outputs: Loading FVQATrainDataset... True done splitting Loading FVQATestDataset... Loading glove... Building Model... Segmentation fault. with NCCL background it starts the training but get stuck and doesn’t go further than this :slight_smile:Please don't send emails directly to my mailbox :) Using GitHub issues can help others to know and solve problems. Original Email: Windows don't have NCCL if you can switch to gloo it might do the trick but I have no idea how to do that Per user-direction, the job has been aborted. ------------------------------------------------------- -------------------------------------------------------------------------- mpirun detected that one or more processes exited with non-zero status, thus causing the job to be terminated.Dec 12, 2022 · Here is a good NVIDIA installation guide. Check if you already have an NVIDIA driver with nvidia-smi.. If you already have the NVIDIA drivers correctly installed, install PyTorch from the official source according to your system. RuntimeError: Distributed package doesn't have NCCL built in. To Reproduce. I install pytorch from the source v1.0rc1, getting the config summary as follows:Aug 19, 2022 · Hi, nngg11, I'm not sure if this codebase supports training / testing on windows since I have never tried this before. I only use linux-based systems, and I guess there will be some problems if you run training / testing on windows. Install the libnccl2 package with YUM. Additionally, if you need to compile applications with NCCL , you can install the libnccl-devel package and optionally the libnccl-static package if you intend to link NCCL statically in your application:The distributed package comes with a distributed key-value store, which can be used to share information between processes in the group as well as to initialize the distributed package in torch.distributed.init_process_group () (by explicitly creating the store as an alternative to specifying init_method .)C._ distributed _ c 10 d import ProcessGroupUCC 118 ProcessGroupUCC.__ module __ = "torch.distributed.distributed_c10d" 119 __all__ += ["ProcessGroupUCC"] 120 except ImportError: 121 _UCC_AVAILABLE = False 122 123 logger = logging. getLogger (__name__) 124 global _c10d_error_logger 125 _c10d_error_logger = _get_or_create_logger 126 127 PG ... 372 raise RuntimeError("Distributed package doesn't have NCCL " 373 "built in" ) 374 _default_pg = ProcessGroupNCCL(store, rank, world_size)Distributed package doesn’t have NCCL built in Hi @nguyenngocdat1995 , sorry for the delay - Jetson doesn’t have NCCL, as this library is intended for multi-node servers. You may need to disable the multiprocessing in the detectron’s training.pytorchlighting报错：raise RuntimeError(“Distributed package doesn‘t have NCCL “RuntimeError: Distribu，代码先锋网，一个为软件开发程序员提供代码片段和技术文章聚合的网站。Nov 2, 2018 · RuntimeError: Distributed package doesn’t have NCCL built in I install pytorch from the source v1.0rc1, getting the config summary as follows: USE_NCCL is On, Private Dependencies does not include nccl, nccl is not built-in. Aug 19, 2022 · RuntimeError: Distributed package doesn't have NCCL built in #5. RuntimeError: Distributed package doesn't have NCCL built in. #5. Closed. AIisCool opened this issue on Aug 19, 2022 · 1 comment. qiuzhongwei-USTB closed this as completed on Dec 13, 2022. Sign up for free to join this conversation on GitHub . edited. Install CUDA's latest toolkit 10.1 and equivalent CuDNN 7.5.1. Install Openmpi v3.1.2 with CUDA support. Build / install pytroch from source. Test any communication for a process group with mpi backend. PyTorch Version (e.g., 1.0): 1.1. OS (e.g., Linux): Ubuntu 16.04. How you installed PyTorch ( conda, pip, source): installed from ...[Solved] Pyinstaller Package and Run Error: RuntimeError: Unable to open/read ui device Just made a Python program to calculate body mass index BMI, and used Pyside6 to draw the user interface. When using auto-py-exe ( auto-py-to-exe is based on pyinstaller, compared to pyinstaller, it has more GUI interface, which makes it easier to use. for ...pytorchlighting报错：raise RuntimeError(“Distributed package doesn‘t have NCCL “RuntimeError: Distribu，代码先锋网，一个为软件开发程序员提供代码片段和技术文章聚合的网站。 Aug 9, 2021 · How to train a custom model under Windows 10 with miniconda? Inference works great but when I try to start a custom training only errors come up. Latest RTX/Quadro driver and Nvida Cuda Toolkit 11.3 + cudnn 11.3 + ms vs buildtools are in... Hi, i try to run train.py in Windows. Help me please solve the problem. System parameters 12th Gen Intel(R) Core(TM) i5-12600KF 3.70 GHz 32 GB Cuda 11.8 Windows 11 Pro Python 3.10.11 Command: torch... RuntimeError: Distributed package doesn't have NCCL built in. distributed. 23: 8639: August 22, 2023 ← previous page next page ...Dec 3, 2020 · The multiprocessing and distributed confusing me a lot when I’m reading some code. #the main function to enter def main_worker(rank,cfg): trainer=Train(rank,cfg) if __name__=='_main__': torch.mp.spawn(main_worker,nprocs=cfg.gpus,args=(cfg,)) #here is a slice of Train class class Train(): def __init__(self,rank,cfg): #nothing special if cfg.dist: #forget the indent problem cause I can't make ... Distributed package doesn’t have NCCL built in Hi @nguyenngocdat1995 , sorry for the delay - Jetson doesn’t have NCCL, as this library is intended for multi-node servers. You may need to disable the multiprocessing in the detectron’s training.pytorchlighting报错：raise RuntimeError(“Distributed package doesn‘t have NCCL “RuntimeError: Distribu，代码先锋网，一个为软件开发程序员提供代码片段和技术文章聚合的网站。 Start multiple jobs on one computer. You need to specify a different port for each job (29500 by default) to avoid communication conflict. the solution is to specify the port while running the program, and give the port number arbitrarily before the PY file to be executed: python -m torch.distributed.launch --nproc_per_node=1 --master_port ...Aug 19, 2023 · System Info PyTorch version : 2.01 and nightly NVIDIA-SMI 535.54.03 Driver Version: 535.54.03 CUDA Version: 12.2 I installed cuda 11.8 with conda by pip install -r requirements.txt . Ubuntu 2204 wi... raise RuntimeError("Distributed package doesn't have NCCL "RuntimeError: Distributed package doesn't have NCCL built in. And when I print following option in python ...Distributed package doesn't have NCCL? #33. Closed. ericnograles opened this issue on Mar 29 · 2 comments.
Aug 19, 2022 · RuntimeError: Distributed package doesn't have NCCL built in #5. RuntimeError: Distributed package doesn't have NCCL built in. #5. Closed. AIisCool opened this issue on Aug 19, 2022 · 1 comment. qiuzhongwei-USTB closed this as completed on Dec 13, 2022. Sign up for free to join this conversation on GitHub . . Who owns ka
Hi @Anastassia Kornilova Great to meet you, and thanks for your question! Let's see if your peers in the community have an answer to your question.To rebuild or reinstall the package, you can follow the directions in the documentation of the relevant framework. Verify GPU drivers: Ensure your computer has the necessary GPU drivers installed. For NCCL to work appropriately, suitable GPU drivers are needed.raise RuntimeError("Distributed package doesn't have NCCL " "built in") RuntimeError: Distributed package doesn't have NCCL built in Any help would be greatly appreciated, and I have no problem compensating anyone who can help me solve this issue. Aug 9, 2021 · How to train a custom model under Windows 10 with miniconda? Inference works great but when I try to start a custom training only errors come up. Latest RTX/Quadro driver and Nvida Cuda Toolkit 11.3 + cudnn 11.3 + ms vs buildtools are in... Jun 19, 2023 · Hi @anastassia_kor1,. For CPU-only training, TrainingArguments has a no_cuda flag that should be set. For transformers==4.26.1 (MLR 13.0) and transformers==4.28.1 (MLR 13.1), there's an additional xpu_backend argument that needs to be set as well. Saved searches Use saved searches to filter your results more quicklyHere is a good NVIDIA installation guide. Check if you already have an NVIDIA driver with nvidia-smi.. If you already have the NVIDIA drivers correctly installed, install PyTorch from the official source according to your system.Mar 25, 2021 · RuntimeError: Distributed package doesn’t have NCCL built in All these errors are raised when the init_process_group () function is called as following: torch.distributed.init_process_group (backend='nccl', init_method=args.dist_url, world_size=args.world_size, rank=args.rank) Here, note that args.world_size=1 and rank=args.rank=0. Hi, i try to run train.py in Windows. Help me please solve the problem. System parameters 12th Gen Intel(R) Core(TM) i5-12600KF 3.70 GHz 32 GB Cuda 11.8 Windows 11 Pro Python 3.10.11 Command: torch... System Info PyTorch version : 2.01 and nightly NVIDIA-SMI 535.54.03 Driver Version: 535.54.03 CUDA Version: 12.2 I installed cuda 11.8 with conda by pip install -r requirements.txt . Ubuntu 2204 wi...Dec 8, 2021 · raise RuntimeError("Distributed package doesn't have NCCL "RuntimeError: Distributed package doesn't have NCCL built in. And when I print following option in python ... You signed in with another tab or window. Reload to refresh your session. You signed out in another tab or window. Reload to refresh your session. You switched accounts on another tab or window.RuntimeError: Distributed package doesn’t have NCCL built in All these errors are raised when the init_process_group () function is called as following: torch.distributed.init_process_group (backend='nccl', init_method=args.dist_url, world_size=args.world_size, rank=args.rank) Here, note that args.world_size=1 and rank=args.rank=0.dist_util.setup_dist()---> RuntimeError: Distributed package doesn't have NCCL built in 👍 3 nathanterroir, kbatsuren, and TneitaP reacted with thumbs up emoji All reactionsraise RuntimeError(“Distributed package doesn‘t have NCCL “ “built in“) RuntimeError: Distributed pa_lanmy_dl的博客-程序员秘密. 技术标签： 训练过程 安装配置 python ubuntu pytorch 服务器The torch.distributed package also provides a launch utility in torch.distributed.launch. This helper utility can be used to launch multiple processes per node for distributed training. torch.distributed.launch is a module that spawns up multiple distributed training processes on each of the training nodes. RuntimeError: Distributed package doesn't have NCCL built in #722. Open jclega opened this issue Aug 26, 2023 · 0 comments Open RuntimeError: Distributed package ... 正文 我和宋清朗相恋三年，在试婚纱的时候发现自己被绿了。 大学时的朋友给我发了我未婚夫和他白月光在一起吃饭的照片。.