MVP Surrogate Plugin ML/AI

[franflame]

#1158

Summary/Motivation:

Added MVP surrogate plugin for creating ML/AI models

Instructions to run new plugins, applicable to all three: keras_nn, pytorch_nn, and scikit_nn:
(optional steps are marked with *)

• Click on home button/"Session" and select "Open Session..."

• Open the file FOQUS/examples/tutorial_files/Flowsheets/Tutorial_4/Simple_flow.foqus

• Click on "Surrogate"

• Under the "Tool:" dropdown, select either keras_nn, pytorch_nn, or scikit_nn
  *If you go to the "Data" tab, you should see that the "Flowsheet Results" have been populated

• Go to the tab "Variables" to select your desired input and output variables
  *You can change various settings with "Method Settings"

• Press the green play button to start the training process

• Go to the "Execution" tab to see the output of the plugin

Legal Acknowledgement

By contributing to this software project, I agree to the following terms and conditions for my contribution:

1. I agree my contributions are submitted under the copyright and license terms described in the LICENSE.md file at the top level of this directory.
2. I represent I am authorized to make the contributions and grant the license. If my employer has rights to intellectual property that includes these contributions, I represent that I have received permission to make contributions and grant the required license on behalf of that employer.

[codecov]

Codecov Report

Attention: 40 lines in your changes are missing coverage. Please review.

Comparison is base (cb938bf) 37.67% compared to head (39a1d6f) 38.52%.

Files	Patch %	Lines
foqus_lib/framework/surrogate/scikit_nn.py	83.06%	16 Missing and 5 partials ⚠️
foqus_lib/framework/surrogate/pytorch_nn.py	92.81%	8 Missing and 4 partials ⚠️
foqus_lib/framework/surrogate/keras_nn.py	94.81%	7 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1163      +/-   ##
==========================================
+ Coverage   37.67%   38.52%   +0.85%     
==========================================
  Files         160      163       +3     
  Lines       36439    36880     +441     
  Branches     5961     5995      +34     
==========================================
+ Hits        13727    14208     +481     
+ Misses      21609    21562      -47     
- Partials     1103     1110       +7     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[bpaul4]

@franflame this is very nice! I tested it locally and it ran smoothly. Just a few minor comments, overall the MVP looks good.

[bpaul4]

@franflame @lbianchi-lbl I just tested this by manually executing the "Simple_flow" test in the GUI, checking the save locations and loading behavior of the saved model files, and fixing a few typos to make the Spell Checker happy. It looks very nice, great work! The .keras extension is different than the .h5 extension we use in the ML AI Plugin but seems to work properly. The Plugin currently assumes file extensions directly correspond to model types and doesn't account for same-name models of different extensions, so we should probably update that in a separate PR.

I wasn't able to add samples after loading from a CSV file. Loading from a flowsheet node worked great. Would there be a way to use loaded data that does exist on the flowsheet to generate samples that the "Variables" tab can recognize? The UQ module may give some inspiration for how to do this.

[bpaul4]

Everything is good, GUI and data improvements planned for future updates.