Skip to content

Conversation

@Oelgazz
Copy link
Contributor

@Oelgazz Oelgazz commented May 11, 2024

closes #371

@Oelgazz Oelgazz added enhancement New feature or request good first issue Good for newcomers refactor doesn't change functionality, just improves code labels May 11, 2024
@Oelgazz Oelgazz requested a review from jardiamj May 11, 2024 16:58
@Oelgazz Oelgazz self-assigned this May 11, 2024
@jardiamj
Copy link
Contributor

This looks good. There are just some empty lines and 4-space indentations that clang-format complains about in the C++ code and one small suggestion for the Python script.

@jardiamj
Copy link
Contributor

I have one question about the purpose of the Python script. Is it meant to be a tool for generating the neural network graph, with x and y locations, for a given grid size (e.g. 10x10, 100x100)? if so, it might be more useful for the script to accept the grid size, through the command line for example.

@Oelgazz
Copy link
Contributor Author

Oelgazz commented May 19, 2024

The purpose of the python script is to be a tool for users to generate GraphML files for the neural simulator. This script is the initial version and is updated in other branches to include parameters such as inhibitory and endogenously active neuron lists (see issues #641 and #642 ).

Because of this, it will be difficult for a user to input these details through the command line, as they will need to at least include the grid size, the inhibitory neuron list, the endogenously active neuron list, and the graph name. Also, since using GraphML will allow the possibility of having non-grid layouts, the grid size may become a (x,y) tuple list.

@jardiamj
Copy link
Contributor

The purpose of the python script is to be a tool for users to generate GraphML files for the neural simulator. This script is the initial version and is updated in other branches to include parameters such as inhibitory and endogenously active neuron lists (see issues #641 and #642 ).

Because of this, it will be difficult for a user to input these details through the command line, as they will need to at least include the grid size, the inhibitory neuron list, the endogenously active neuron list, and the graph name. Also, since using GraphML will allow the possibility of having non-grid layouts, the grid size may become a (x,y) tuple list.

Great! That makes sense.

@stiber
Copy link
Contributor

stiber commented May 23, 2024

I'll create an issue relating to the above for future work. We actually have an algorithm to create the pattern of neuron types from the grid size, which is embedded in the Workbench neuron layout editor code (NLEdit). This would allow the tool to just take in something like the grid size and compute all of the other stuff.

@Oelgazz Oelgazz merged commit af23229 into development May 31, 2024
@Oelgazz Oelgazz deleted the issue-371-refactor-neural-init branch May 31, 2024 19:27
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

enhancement New feature or request good first issue Good for newcomers refactor doesn't change functionality, just improves code

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants