Skip to content

NN Train Rework for Customizability#12

Open
Ahaan1234 wants to merge 46 commits intomainfrom
nn_train-rework
Open

NN Train Rework for Customizability#12
Ahaan1234 wants to merge 46 commits intomainfrom
nn_train-rework

Conversation

@Ahaan1234
Copy link
Collaborator

Added two runtime arguments, --feature_cols and --conf_thresh, that allow you to select certain feature columns and confidence thresholds for training the neural network.

default for feature columns is "Envelope H1,Envelope H2,Envelope H3", and you can select any columns (or number) in the format "Col X,Col Y,Col Z,…" Tested with single GCC columns, 2 GCC column, and all Envelope columns

default for confidence thresholds is "0.5,0.8" (decimal value of desired percentage). You can use any number (between 0 and 1) and any set any amount of thresholds. Tested with 2, 3 and 4 confidence thresholds

JaredBaileyDuke and others added 30 commits October 19, 2025 14:02
Read binary from python, Refactor Code
Ignore standard files, virtual environments, data, and model artifacts
Revert "Read binary from python, Refactor Code"
Updated cpp read script to include selection of bin files with ml arg
…version, simplified read function, leaving commented code in case of future errors
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants