Skip to content

Latest commit

 

History

History
31 lines (21 loc) · 1.95 KB

notes.md

File metadata and controls

31 lines (21 loc) · 1.95 KB

  1. For Fortran compiler: -fdefault-real-8 -fdefault-double-8

  2. When writting Python code and generate a new file. Always close a file after using it.

  3. use zip -r folder.zip folder, and unzip folder to compress and uncompres files

  4. Long range interaction might be importance for CO2 hydrate clathrate.

  5. Bas's Library fitting routine is weighted while in MSA ther is no weight. The benefit of weighting in Bas's library is that the extremely big data won't affact too much.

  6. Molpro could result in problem such as GLOBAL ERROR fehler on processor 0

  7. About compiler: ifort is Intel's commercial compiler. mkl is Intel's math kernal library.(This one can be downloaded for free)--https://software.intel.com/en-us/intel-mkl mkl is located at cd /opt/intel/mkl/

  8. PGI, pgf90 is another option for compiler. https://www.pgroup.com/purchase/freepgi.php You need to download a lisence to the /opt/pgi/ as lisence.dat file. A developer free version lisence can be obtained through NVIDIA OpenACC Toolkit programe. Ater installation and activation, you can open configured terminal in /opt/pgi/, for example, open pgdbg64.app. Then you can use commands such as pgf90 to complie your Fortran codes. Or, as Mac user, you can add the following command in the Terminal>>Preference>>Profiles>>(your default terinal)>>Shell>>Startup, check Run command and Run inside shell, and enter: setenv PATH /opt/pgi/osx86-64/2016/bin:/opt/pgi/osx86-64/2016/mpi/mpich/bin:$PATH;export PATH=/opt/pgi/osx86-64/2016/bin:/opt/pgi/osx86-64/2016/mpi/mpich/bin:$PATH; clear

  9. Use printenv to print global variable.

  10. Use tar -vzvf 'test.tar' * to compress all current files. tar -czvf 'test.tar.bz2' * is even better. In order to unzip: tar -xzvf archive.tar.gz

  11. Use grep -r 'string' ./ to search for string in the all files in current folder.

  12. Simply fit the feature if you have only one feature. But for a 4000 * 904 matrix, you have to think about it.