Failure to install GLIF module

[johnhongyumeng]

Dear Contributors,

I'm trying to work on the Allen visual cortex model, which requires the glif module. I'm using Ubuntu 20.02 system and I installed the latest version of NEST by anaconda with python 3.8. I have installed libtools.

However, when I try to use the following command to install glifmodule

cmake --Dwith-nest=nest-config -Dwith-ltdl=ON [-Dwith-mpi=ON] ../GlifModel

as mentioned in the installation page

https://github.com/AllenInstitute/GLIF2NEST

I got the following error:

-- The CXX compiler identification is unknown CMake Error at CMakeLists.txt:75 (project): The CMAKE_CXX_COMPILER:

/home/conda/feedstock_root/build_artifacts/nest-simulator_1583214391390/_build_env/bin/x86_64-conda_cos6-linux-gnu-c++

is not a full path to an existing compiler tool.

Tell CMake where to find the compiler by setting either the environment variable "CXX" or the CMake cache entry CMAKE_CXX_COMPILER to the full path to the compiler, or to the compiler name if it is in the PATH.

-- Configuring incomplete, errors occurred! See also "/home/yawntaiyuan/GLIF2NEST/build/CMakeFiles/CMakeOutput.log". See also "/home/yawntaiyuan/GLIF2NEST/build/CMakeFiles/CMakeError.log".

Could you help me with the problem?

More information: I have tested the CMake, Compiler, NEST simulator. They all works properly by themselves. I have updated all the packages to the newest.

I have checked the passage in the error message: /home/conda/feedstock_root/build_artifacts/nest-simulator_1583214391390/_build_env/bin/x86_64-conda_cos6-linux-gnu-c++

and indeed I don't have that file at the location. However, I did install c++. Is it possible that I should install other packages before as well?

Best, John

I could be reached by john.meng@nyu.edu. I'd like to test some of predictions of our rate model by using a point-neuron model. If anyone like to help on this project, I would deeply appreciate it.

[jougs]

Is there a specific reason you're compiling the module yourself, instead of relying on the most current version of the models that are built into NEST (as the GLIF2NEST README suggests)? See #1126 for the PR in which they got merged.

[johnhongyumeng]

Dear Jochen,

Thanks for replying!

I am trying to run a model for V1 in mice, developed by Allen Institute. The description of the model is in https://portal.brain-map.org/explore/models/mv1-all-layers.

The model is based on the GLIF model. I downloaded their data and run the code based on their description in https://www.dropbox.com/sh/w5u31m3hq6u2x5m/AACpYpeWnm6s_qJDpmgrYgP7a?dl=0&preview=README.txt

When I run the following command as they suggested

mpirun -np 1 python run_pointnet.py config.json

I got the following error

Traceback (most recent call last): File "run_pointnet.py", line 9, in nest.Install('glifmodule') File "/opt/nest/lib/python3.8/site-packages/nest/ll_api.py", line 246, in stack_checker_func return f(*args, **kwargs) File "/opt/nest/lib/python3.8/site-packages/nest/lib/hl_api_simulation.py", line 318, in Install return sr("(%s) Install" % module_name) File "/opt/nest/lib/python3.8/site-packages/nest/ll_api.py", line 132, in catching_sli_run raise exceptionCls(commandname, message) nest.lib.hl_api_exceptions.DynamicModuleManagementError: ("DynamicModuleManagementError in Install: Module 'glifmodule' could not be opened.\n The dynamic loader returned the following error: 'file not found'.\n\nPlease check LD_LIBRARY_PATH (OSX: DYLD_LIBRARY_PATH)!", 'DynamicModuleManagementError', , ": Module 'glifmodule' could not be opened.\n The dynamic loader returned the following error: 'file not found'.\n\nPlease check LD_LIBRARY_PATH (OSX: DYLD_LIBRARY_PATH)!")

Thus, I thought the problem was that I hadn't installed the glifmodule. I searched and found the source code. During the installation of the module, I encountered the problem and reported as in https://github.com/AllenInstitute/GLIF2NEST/issues/3

For your information, I installed the NEST by compiling the source code in an anaconda environment, following the blog http://neurochannels.blogspot.com/2019/02/installing-nest-simulator-for-use-with.html The default installation method didn't work for me. The version of NEST is 2.20.0

Do you suggest that I might have installed the glifmodel but I set the environment wrong? I have searched in my lib, but no file named as "glifmodule" is found. It's not listed in the results of $ conda list

Thanks again for your reply! I truly appreciate your help. I have stuck here for quite a while now.

Best regards, John Meng

On Wed, Sep 2, 2020 at 4:37 AM Jochen Martin Eppler < notifications@github.com> wrote:

Is there a specific reason you're compiling the module yourself, instead of relying on the most current version of the models that are built into NEST (as the GLIF2NEST README https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_AllenInstitute_GLIF2NEST-23update&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=p0jOqllLszlxCqWhnhAGspEzB2nk0IaLDbDfQvz8pE0&m=R0N7asI0wH4lfmaAj_yaK8MIvcgjiSA5GYFnerB2G_g&s=7S6XJTY0gUlIpehmV87pVItJlk6EflvEDufNJDaIbII&e= suggests)? See #1126 https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_nest_nest-2Dsimulator_pull_1126&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=p0jOqllLszlxCqWhnhAGspEzB2nk0IaLDbDfQvz8pE0&m=R0N7asI0wH4lfmaAj_yaK8MIvcgjiSA5GYFnerB2G_g&s=yUadCeW2zpe57MibVvULnDXn_ixNzdiJnUaOMMCedxQ&e= for the PR in which they got merged.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_nest_nest-2Dsimulator_issues_1738-23issuecomment-2D685447031&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=p0jOqllLszlxCqWhnhAGspEzB2nk0IaLDbDfQvz8pE0&m=R0N7asI0wH4lfmaAj_yaK8MIvcgjiSA5GYFnerB2G_g&s=NoSEIdLtzaoLfDdJaQfK1jmWgv075VY-PpBxH5295jc&e=, or unsubscribe https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_ANVVLPZ6Z7K2P47WESVR4SDSDX76DANCNFSM4QOQXDEA&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=p0jOqllLszlxCqWhnhAGspEzB2nk0IaLDbDfQvz8pE0&m=R0N7asI0wH4lfmaAj_yaK8MIvcgjiSA5GYFnerB2G_g&s=9VgZdcqb0wcmkSKKFxaptDXtQyRqwvMJeKd5Erlsza8&e= .

[jougs]

What I wanted to say is that you don't need the glifmodule anymore, because the neuron models contained in it have been integrated into NEST proper when #1126 was merged. This means that you must not call nest.Install('glifmodule') in your simulation script, as that module does not exist. The models should nevertheless be available.

[johnhongyumeng]

Thanks so much! I will test that by removing the line. Otherwise I will contact the authors of the model. As soon as I find out a solution, I will close the issue on Github.

Best, John

On Wed, Sep 2, 2020 at 8:00 AM Jochen Martin Eppler < notifications@github.com> wrote:

What I wanted to say is that you don't need the glifmodule anymore, because the neuron models contained in it have been integrated into NEST proper when #1126 https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_nest_nest-2Dsimulator_pull_1126&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=p0jOqllLszlxCqWhnhAGspEzB2nk0IaLDbDfQvz8pE0&m=711qE94HrDkbfeaJsdrCAbfgh_uHAIZ0c4PMkXH5RlE&s=Z81uuNwjgAph-BcNFxrbxGByoq7SwD60ppxfBd7uckU&e= was merged. This means that you must not call nest.Install('glifmodule') in your simulation script, as that module does not exist. The models should nevertheless be available.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_nest_nest-2Dsimulator_issues_1738-23issuecomment-2D685684466&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=p0jOqllLszlxCqWhnhAGspEzB2nk0IaLDbDfQvz8pE0&m=711qE94HrDkbfeaJsdrCAbfgh_uHAIZ0c4PMkXH5RlE&s=d2FiMzuwNH-5aSfDQmCKdQwVwD3vzXBkkmF82zki4Jg&e=, or unsubscribe https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_ANVVLP6YXYVMZAOW3ITJ4XTSDYXU7ANCNFSM4QOQXDEA&d=DwMCaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=p0jOqllLszlxCqWhnhAGspEzB2nk0IaLDbDfQvz8pE0&m=711qE94HrDkbfeaJsdrCAbfgh_uHAIZ0c4PMkXH5RlE&s=7ON6KvO7-fE0d0jk8uors2yWC_7zhaDFsPyzjkbYcu4&e= .

[gtrensch]

Issue closed. Please reopen if this is still a problem.