pip install -v -e . error #22
Comments
|
@scuizhibin Your PyTorch is installed with 12.1. Try Pytorch+cuda11.8 |
|
@scuizhibin Did you solve the problem? I faced same error to you |
|
生成map数据的时候 线程杀死，这个 会是什么原因
…---Original---
From: "Xiaosong ***@***.***>
Date: Thu, Aug 8, 2024 12:45 PM
To: ***@***.***>;
Cc: ***@***.******@***.***>;
Subject: Re: [Thinklab-SJTU/Bench2DriveZoo] pip install -v -e . error (Issue#22)
@scuizhibin Your PyTorch is installed with 12.1. Try Pytorch+cuda11.8
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you were mentioned.Message ID: ***@***.***>
|
|
@scuizhibin 内存不足，np.load不了，20G内存大概是不够用的，想办法把内存调大点，我后面用了160G反正是成功了。 |
|
您好，我有两个问题，一、是否支持win系统，二、是否一定得安装cuda11.8，我安装的11.7，然后运行pip install -v -e .时报错 ninja: build stopped: subcommand failed. |
|
11.8 要不然会出错，并且需要显卡驱动也要更新一下，硬件要求是3090
…---Original---
From: ***@***.***>
Date: Thu, Aug 15, 2024 15:54 PM
To: ***@***.***>;
Cc: ***@***.******@***.***>;
Subject: Re: [Thinklab-SJTU/Bench2DriveZoo] pip install -v -e . error (Issue#22)
您好，我有两个问题，一、是否支持win系统，二、是否一定得安装cuda11.8，我安装的11.7，然后运行pip install -v -e .时报错 ninja: build stopped: subcommand failed.
Traceback (most recent call last):
File "G:\0_anaconda3\envs\b2d_zoo\lib\site-packages\torch\utils\cpp_extension.py", line 1893, in _run_ninja_build
subprocess.run(
File "G:\0_anaconda3\envs\b2d_zoo\lib\subprocess.py", line 516, in run
raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ninja', '-v', '-j', '55']' returned non-zero exit status 1.
这该如何解决呢？
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you were mentioned.Message ID: ***@***.***>
|
|
@TroyZhangFawang 建议linux, cuda version 为11.8 |
|
@scuizhibin @jayyoung0802 您好，我在“pip install -v -e .”时也遇到了同样的问题， |
|
Can you provide the complete terminal log? |
|
@jayyoung0802 您好，这是完整日志： .... error: subprocess-exited-with-error × python setup.py develop did not run successfully. note: This error originates from a subprocess, and is likely not a problem with pip. 我有注意到“#error "You're trying to build PyTorch with a too old version of GCC. We need GCC 9 or later."，但是我当前编译的gcc版本是9.5 |
问题解决了，尝试过gcc9.5.0, 10.5.0, 11.4.0, 结果gcc必须9.4.0才行。由于我是ubuntu22.04，sudo apt-get install gcc-9安装的都是gcc 9.5.0，需要去gcc官网下载gcc 9.4.0的包 然后先编译好9.4.0再去编译mmcv。 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
The detected CUDA version (11.8) mismatches the version that was used to compile PyTorch (12.1). Please make sure to use the same CUDA versions. my torch version torch-2.4.0
The text was updated successfully, but these errors were encountered: