You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When we implement #109 and have more fleshed out examples, it would be nice to be able to have .ipynb files that users can download and run on their own machines.
Description of idea
We should add functionality in app.py to allow users to run SaltProc inside a python environment while maintaining the current capability to run from the command line.
Implementation details
SaltProc is locked into the command line in the App.parse_arguments function, as App.parse_arguments only looks for input from the command line.
I propose the following:
Add a new function in app.py that will handle inputs within a python context OR add control flow to the current App.parse_arguments function that will detect if we are running from the command line or from a python context.
Add global variables or class attributes to app.py (that will be set from either command line input or python context input) that will make accessible the following variables to the App.run function:
n, the number of nodes to use in parallel execution
d, the number of cores to use in parallel execution
i, the path to the SaltProc input file
use these variables in App.run
Optional
rename app to saltproc for clarity.
Potential snags
It is unclear if/how this will effect performance.
The text was updated successfully, but these errors were encountered:
Background and motivation
When we implement #109 and have more fleshed out examples, it would be nice to be able to have
.ipynb
files that users can download and run on their own machines.Description of idea
We should add functionality in
app.py
to allow users to run SaltProc inside a python environment while maintaining the current capability to run from the command line.Implementation details
SaltProc is locked into the command line in the
App.parse_arguments
function, asApp.parse_arguments
only looks for input from the command line.I propose the following:
app.py
that will handle inputs within a python context OR add control flow to the currentApp.parse_arguments
function that will detect if we are running from the command line or from a python context.app.py
(that will be set from either command line input or python context input) that will make accessible the following variables to theApp.run
function:n
, the number of nodes to use in parallel executiond
, the number of cores to use in parallel executioni
, the path to the SaltProc input fileApp.run
Optional
app
tosaltproc
for clarity.Potential snags
It is unclear if/how this will effect performance.
The text was updated successfully, but these errors were encountered: