ml.mlp

Functions for the multi-layer perceptron classifier.

Functions

build_dataloaders(data_split)	A function to build the DataLoaders from the data split.
create_layers(input_size, n_neurons)
evaluate_model(feature, mlp_cls, ...)	A function to evaluate the model on test data.
format_plots()	General plotting parameters for the Kulik Lab.
gradient_step(model, dataloader, optimizer, ...)	A function to train on the entire dataset for one epoch.
load_data(mimos, include_esp, data_loc)	Load data from CSV files for each mimo in the given list.
optuna_mlp(data_split_type, include_esp, ...)
plot_confusion_matrices(cms, mimos)	Plot confusion matrices for distance and charge features.
plot_data(df_charge, df_dist, mimos)	Plot the average charge and distance data for the given MIMO types.
plot_roc_curve(y_true, y_pred_proba, mimos, ...)	Plot the ROC curve for the test data of the charge and distance features.
plot_train_val_losses(train_loss_per_epoch, ...)	Plot the train and validation losses as a function of epoch number.
preprocess_data(df_charge, df_dist, mimos, ...)	Split train and test based on the given test and validation fractions.
run_mlp(data_split_type, include_esp, ...)
shap_analysis(mlp_cls, train_loader, ...)	Plot SHAP dot plots for each mimichrome to identify importance
train(feature, layers, lr, n_epochs, l2, ...)	A function to train and validate the model over all epochs.
train_with_hyperparameters(trial, feature, ...)
validate(model, dataloader, device)	A function to validate on the validation dataset for one epoch.

Classes

MDDataset(X, y)
MimoMLP(layers)