ElofssonLab / PconsC4

Fast and easy contact prediction.
GNU General Public License v3.0
35 stars 10 forks source link

install failed #7

Open mayu1995-zj opened 4 years ago

mayu1995-zj commented 4 years ago

hello, i install pconsc4 on Python3.6 and tensorflow=1.10.0, but there are some problems: error: Command "gcc -pthread -B /opt/conda/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -DENABLE_PYTHON_MODULE -DPYTHRAN=3 -DPYTHRAN_BLAS_BLAS -I /home/yuma/Pconsc4_venv/lib/python3.6/site-packages/pythran -I/home/yuma/Pconsc4_venv/lib/python3.6/site-packages/numpy/core/include -I/opt/conda/include/python3.6m -c src/gaussdca/_gdca.cpp -o build/temp.linux-x86_64-3.6/src/gaussdca/_gdca.o -std=c++11 -fno-math-errno -w -fvisibility=hidden -fno-wrapv -fopenmp -ftree-vectorize" failed with exit status 1 how to solve it? Thanks.

ghost commented 4 years ago

Hello,

I am also getting this error when trying to install. Please advise!

PuYuQian commented 4 years ago

I am also getting the same error when trying to install. How to solve it?

Dapid commented 4 years ago

Hi,

There is a problem in pythran 0.9.6, but if you revert back to 0.9.5 it should work.

Dapid commented 4 years ago

I suspect some library is missing. Can you please show the full error?

On Tue, 15 Sep 2020, 3:50 am Sulebork, notifications@github.com wrote:

I am also getting a compile error with python35 & pythran 0.9.5: error: Command "gcc -pthread -shared -L/home/fanganpai/.conda/envs/py3.5/lib -B /home/fanganpai/.conda/envs/py3.5/compiler_compat -Wl,-rpath=/home/fanganpai/.conda/envs/py3.5/lib,--no-as-needed build/temp.linux-x86_64-3.5/pconsc4/parsing/_load_data.o -o build/lib.linux-x8664-3.5/pconsc4/parsing/ load_data.cpython-35m-x86_64-linux-gnu.so -O2 -march=native -pipe -mtune=native" failed with exit status 1 error: Command "g++ -pthread -shared -L/home/fanganpai/.conda/envs/py3.5/lib -B /home/fanganpai/.conda/envs/py3.5/compiler_compat -Wl,-rpath=/home/fanganpai/.conda/envs/py3.5/lib,--no-as-needed build/temp.linux-x86_64-3.5/src/gaussdca/_load_data.o -L/usr/lib64 -lblas -lblas -o build/lib.linux-x8664-3.5/gaussdca/ load_data.cpython-35m-x86_64-linux-gnu.so -fvisibility=hidden -Wl,-strip-all" failed with exit status 1 Compiler: g++ (GCC) 10.2.0 & GNU ld (GNU Binutils) 2.35 Does anyone know where the problem is? Thanks.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/ElofssonLab/PconsC4/issues/7#issuecomment-692411869, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAMKYLR6FIZB2H2OA2YIT2DSF3B5HANCNFSM4NBGXTHA .

fanganpai commented 4 years ago

I suspect some library is missing. Can you please show the full error? On Tue, 15 Sep 2020, 3:50 am Sulebork, @.***> wrote: I am also getting a compile error with python35 & pythran 0.9.5: error: Command "gcc -pthread -shared -L/home/fanganpai/.conda/envs/py3.5/lib -B /home/fanganpai/.conda/envs/py3.5/compiler_compat -Wl,-rpath=/home/fanganpai/.conda/envs/py3.5/lib,--no-as-needed build/temp.linux-x86_64-3.5/pconsc4/parsing/_load_data.o -o build/lib.linux-x8664-3.5/pconsc4/parsing/ load_data.cpython-35m-x86_64-linux-gnu.so -O2 -march=native -pipe -mtune=native" failed with exit status 1 error: Command "g++ -pthread -shared -L/home/fanganpai/.conda/envs/py3.5/lib -B /home/fanganpai/.conda/envs/py3.5/compiler_compat -Wl,-rpath=/home/fanganpai/.conda/envs/py3.5/lib,--no-as-needed build/temp.linux-x86_64-3.5/src/gaussdca/_load_data.o -L/usr/lib64 -lblas -lblas -o build/lib.linux-x8664-3.5/gaussdca/ load_data.cpython-35m-x86_64-linux-gnu.so -fvisibility=hidden -Wl,-strip-all" failed with exit status 1 Compiler: g++ (GCC) 10.2.0 & GNU ld (GNU Binutils) 2.35 Does anyone know where the problem is? Thanks. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#7 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAMKYLR6FIZB2H2OA2YIT2DSF3B5HANCNFSM4NBGXTHA .

Sorry for bother , but actually the problem is caused by something else.Downgrading binutils can solve it. I deleted the reply after realized what happened, but somehow it was still visible for you.

yzxu-hzau commented 4 years ago

The requirements of this porject should be more specific, after I tried lots of method to fix the compiling error. I found the solution all requirements: gcc5, pythran==0.9.5, keras=2.2.4, tensorflow=1.14.0. More importantly, beniget==0.2.1, gast=0.3.3. After all that, I recommend users of pip add the new feature 'pip install package --use-feature=2020-resolver'.

PuYuQian commented 4 years ago

gcc version 7.5.0 and pytorch=1.6 can also success. More importantly, beniget==0.2.1, gast=0.3.3

Ruheng-W commented 2 years ago

Recently, i have also encountered this problem and solved it, now i want to share some experience.

  1. When gcc < 5 the pyGaussDCA will not be installed successfully. If hard to update gcc, here i recommend installing a virtual environment by using devtoolset.
  2. When tensorflow >= 2 the pconsc may be installed successfully, but it can not be used in python file. So the version of tensorflow must < 2.
  3. As for python2 or python3, here i use the python3.7 and install the pconsc4 successfully, but it may occur some problems because of the conflict with the version of keras, but it will be solved eazily. I also test the python2.7, but i can not install pconsc4 and PYG under this condition, so finally i choose the version 3.7 of python. For pytorch, choosing anyone version that fit your os and the version of python you installed is ok.
  4. So i recomend the following requirements: gcc7.5, pythran==0.9.5, keras=2.2.4, tensorflow=1.14.0, python3.7, and pytorch 1.7. Thanks for reading, and hoping these tips will help you!
calum-hand commented 2 years ago

See proposed PR #10

Plovor commented 2 years ago

I failed by gcc-4.8, and gcc-5.1 works. beniget-0.2 and gast-0.3.3 are also important.

daydayupzzl commented 2 years ago

gcc version 7.5.0 and pytorch=1.6 can also success. More importantly, beniget==0.2.1, gast=0.3.3

非常感谢~ 解决问题,成功安装!