The end-user algorithms are currently scattered around; some are in the example folder of the main branch, others are in new repositories such as https://github.com/Cytnx-dev/Cytnx_examples and https://github.com/Cytnx-dev/TRG
It would be great to move them all to a central place, such that users can find and use them easily. This can be the main repository, but I suggest having a new algorithm repository to separate the core functionalities from end-user algorithms (but this can be discussed).
What would be needed for each algorithm to be moved to the new location:
- update to work with the newest version of Cytnx
- clean up the code and make the coding style uniform
- write unit tests
- write API documentation and make it available online
- write user doc and make it available online
- write usage examples (but there is overlap in this with unit tests and user guide)
An example to start with could be the DMRG algorithm. There are many versions out there (different folders online, summer school tutorial, I have my own). While it might be good to have different implementations for different purposes, it would be good to support one version officially.
This is related to #739
The end-user algorithms are currently scattered around; some are in the example folder of the main branch, others are in new repositories such as https://github.com/Cytnx-dev/Cytnx_examples and https://github.com/Cytnx-dev/TRG
It would be great to move them all to a central place, such that users can find and use them easily. This can be the main repository, but I suggest having a new algorithm repository to separate the core functionalities from end-user algorithms (but this can be discussed).
What would be needed for each algorithm to be moved to the new location:
An example to start with could be the DMRG algorithm. There are many versions out there (different folders online, summer school tutorial, I have my own). While it might be good to have different implementations for different purposes, it would be good to support one version officially.
This is related to #739