forked from NathanKlineInstitute/SMARTAgent
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathinstall.txt
94 lines (78 loc) · 3.15 KB
/
install.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
=============================================================
condensed commands:
if using miniconda:
bash ~/Downloads/Miniconda3-latest-Linux-x86_64.sh
if using anaconda:
bash ~/Downloads/Anaconda3-2020.02-Linux-x86_64.sh
conda install numpy scipy matplotlib pandas scikit-image
pip install 'gym[atari]'
conda install imageio-ffmpeg -c conda-forge
conda upgrade ffmpeg
conda install gcc_linux-64 openmpi mpi4py cython
conda install gxx_linux-64
pip install neuron-nightly
pip install netpyne --upgrade
cd ~/SMARTAgent
nrnivmodl mod
=============================================================
on cycle (server at nki)
as root installed full anaconda in /opt/anaconda3
using
bash ~/Downloads/Anaconda3-2020.02-Linux-x86_64.sh
then created lab group and set group for /opt/anaconda3 to lab
and owner to samn
make sure to export LD_LIBRARY_PATH=/opt/anaconda3/lib
also have to use pipz instead of pip on nki machines
pipz alias is alias pipz='pip --trusted-host files.pythonhosted.org --trusted-host pypi.org --trusted-host pypi.python.org'
and make sure those aliases and exports in .bashrc
conda install numpy scipy matplotlib pandas scikit-image
pipz install 'gym[atari]'
conda install imageio-ffmpeg -c conda-forge
conda upgrade ffmpeg
conda install gcc_linux-64 openmpi mpi4py cython
conda install gxx_linux-64
pipz install neuron-nightly
pipz install netpyne --upgrade
cd ~/SMARTAgent
nrnivmodl mod
=============================================================
some of the commands with explanations
first need to install anaconda or miniconda (to save space)
downloaded and ran Miniconda3-latest-Linux-x86_64.sh
installed in /home/samn/miniconda3
bash ~/Downloads/Miniconda3-latest-Linux-x86_64.sh
then some python packages via conda and pip
conda install numpy scipy matplotlib pandas scikit-image
pip install 'gym[atari]'
conda install imageio-ffmpeg -c conda-forge
conda upgrade ffmpeg
this is easier way to install openmpi and other requirements too:
conda install gcc_linux-64 openmpi mpi4py cython
echo $PATH
/home/samn/miniconda3/bin:/home/samn/miniconda3/condabin:/home/samn/bin:/home/samn/.local/bin:/home/samn/bin:/home/samn/.local/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games
export PATH="/home/samn/openmpi/bin:$PATH"
and put that into .bashrc
pip install neuron-nightly
pip install netpyne --upgrade
then from terminal in the SMARTAgent folder:
nrnivmodl mod
to compile the NEURON mod files
=================================================================================================================
for gcp with ubuntu 20 had to also install ...
zlib1g-dev, build-essential, cmake, pkg-config (for matplotlib in netpyne) and some others ...
sudo apt-get install zlib1g-dev
so that could install atarai[gym]
sudo apt-get install build-essential
sudo apt-get install cmake
sudo apt-get install pkg-config
sudo apt-get install libc6-dev
sudo apt-get install flex
sudo apt-get install binutils-dev
libelf-dev
libdwarf-dev
still had problems with pip install netpyne --upgrade
so had to install it from git repo development branch
git clone git@github.com:Neurosim-lab/netpyne.git
cd netpyne
git checkout development
pip install -e .