Configuring analysis machine: Difference between revisions

From MEG Core
Jump to navigation Jump to search
Content added Content deleted
Line 2: Line 2:
====Installing Pyctf====
====Installing Pyctf====
Download the pyctf folder
Download the pyctf folder
git clone -devel $USER@$GITSERVER:pyctf
git clone $USER@$GITSERVER:pyctf
cd $PYCTF_FOLDER_PATH
cd $PYCTF_FOLDER_PATH
make
make
make symlinks
make symlinks
#This may error out if you have the mkl library installed >> you can create another conda environment conda creat -n compile python=3.7 nomkl numpy >> conda activate compile >> make >> make symlinks
Add the pyctf path to your pythonpath variable
Add the pyctf path to your pythonpath variable

Revision as of 13:11, 19 February 2020

Under Construction

Installing Pyctf

Download the pyctf folder

 git clone $USER@$GITSERVER:pyctf
 cd $PYCTF_FOLDER_PATH
 make 
 make symlinks 
 #This may error out if you have the mkl library installed >> you can create another conda environment conda creat -n compile python=3.7 nomkl numpy >> conda activate compile >> make >> make symlinks
 

Add the pyctf path to your pythonpath variable

 PYTHONPATH=$PYTHONPATH:$PYCTF_FOLDER_PATH

If you are using an Anaconda python manager

  • Create a file pyctf.pth
 echo $PYCTF_FOLDER_PATH > pyctf.pth
  • Add at pyctf_pth file to your Anaconda folder
 For example /home/User/miniconda3/envs/ENV_NAME/lib/python3.7/site-packages/pyctf.pth

Installing SAMsrc

 #Install gsl
 #Fedora
 sudo dnf install nifticlib-devel
 sudo dnf install gifticlib-devel
 sudo dnf install motif motif-devel
 sudo dnf install fftw fftw-devel
 sudo dnf install libpng12.x86_64
 sudo dnf install -y tcsh libXp op  ## Not sure of the op
 sudo dnf install gsl, gsl-devel    ## If errors when running SAM software, you may need to link the library

Linking thd_atr

 #Afni related functions
 ln -s $SAM/lib/thd_atr $PYCTF_dir/thd_atr

Installing Afni

https://afni.nimh.nih.gov/pub/dist/doc/htmldoc/background_install/install_instructs/index.html

Installing CTF software

To view and process the data using the native ctf software, the packages have been incorporated into a singularity container
Install singularity on your computer

 #Fedora Linux
 sudo dnf install singularity
 #Redhat/Centos Linux
 sudo yum install singularity
 #Ubuntu/Debian Linux 
 sudo apt-get install singularity-container

Download the current CTF singularity container and components

  • ctf.sif
  • wrap
  • mkwrappers
 cd $Directory_With_CTF_Container
 mkwrappers  #This will install the container in /usr/local/ctf/bin and the links to all of the functions in /usr/local/ctf/bin

Environmental Variables that may need to be modified

  • CTF_DATADIR
  • CTF_MRIDIR
  • CTF_WORKDIR

Enabling jupyter lab widgets

 dnf install nodejs
 conda install -c conda-forge ipywidgets
 jupyter nbextension enable --py widgetsnbextension
 jupyter labextension install @jupyter-widgets/jupyterlab-manager