Python_Lib | 使用 Python 库的代码记录 - 这是一个存放使用过的 Python 库的简单示例仓库，欢迎交流 | Data Manipulation library

how do I set Debug vs Release, and similar configuration options?

There are two common options. A plain if:

Downgrading to conda 4.5.5 fixed this for me.

And with a ranting hat on: This is the third time in the space of about 18 months that a conda update has resulted in an error that causes builds to fail, and there's been plenty of other issues too. I presume it will be fixed by the core conda people at some point, and maybe it is caused by a particular combination of third party dependencies, but one way or another conda always seems extremely buggy. There was once an update that caused conda install conda to delete itself!

Thank you @flexo for pointing out the problem. On Windows, we have to make a .pyd, not a .so.

Besides the changes pointed out by @Flexo, there are other things that have to be changed too. The minimal example's Makefile correct version is:

When you change the environment，have you edit the path in Makefile.config?
Please make the python path correct and make sure cudnn is installed，if you want to compile with cuDNN.
Not very clear with your issue, try comment USE_CUDNN := 1

Graph-tool 2.28 requires macOS Mojave. It’s been updated in Homebrew on June, 9. You can try installing the previous version by using a direct URL to the formula at the commit just before the bump:

Goto Makefile (and not the .config file) and change the name of the ld library to python27.

File "/home/breeze/anaconda3/lib/python3.6/site-packages/dateutil/rrule.py", line 55 raise ValueError, "Can't create weekday with n == 0" ^ SyntaxError: invalid syntax

From the error, you understand that you need to update dateutil lib.

I don't have a specific solution. However, when I tripped over these problems, the errors were due to lack of the proper library to link. This came from two causes: (1) library not installed; (2) wrong library version earlier in the LD_LIBRARY_PATH. Unfortunately, I don't recall how my IT support traced and fixed the problem for me -- it took someone who knew the OS library installation much better than my "sudo yum install" level of expertise. :-)

You also commented that changing the compiler worked: another possibility, as a new compiler has a new load path. Good work!

Your /usr/local/lib/libboost_system.so.1.66.0 is evidently one you built yourself and want the loader to find at runtime without special measures.

But having built it, you didn't update the ldconfig cache (because you didn't know you had to); so the runtime loader is not yet aware that this library exists and can't find it.

When the loader is looking for the shared libraries needed to assemble a new process, it does not search all of the linker's default search directories. That would be slow. By default it searches a cached database /etc/ld.so.cache, of libraries that were found by ldconfig, last time it was run.

By default, ldconfig caches libraries that it finds in /lib, /usr/lib and in the directrories listed in the file /etc/ld.so.conf, and/or any similar *.conf files that are recursively include-ed in /etc/ld.so.conf. For example: