hookeai.model_architectures.rc_base_model.train.training

Training of recurrent constitutive model.

Classes

EarlyStopper

Early stopping procedure (implicit regularizaton).

Functions

train_model

Training of recurrent constitutive model.

save_parameters_history

Save model learnable parameters history record.

read_parameters_history_from_file

Read model learnable parameters history from parameters record file.

save_best_parameters

Save best performance state model parameters.

read_best_parameters_from_file

Read best performance state model parameters from file.

check_model_parameters_convergence

Check convergence of model parameters.

Functions

check_model_parameters_convergence(...[, ...])	Check convergence of model parameters.
data_scaler_transform(model, tensor, ...[, mode])	Perform data scaling operation on features PyTorch tensor.
fit_data_scalers(model, dataset[, ...])	Fit model data scalers.
get_learning_rate_scheduler(optimizer, ...)	Get PyTorch optimizer learning rate scheduler.
get_model_summary(model[, input_data, ...])	Get summary of PyTorch model.
get_pytorch_loss(loss_type, **kwargs)	Get PyTorch-based loss function.
get_pytorch_optimizer(algorithm, params, ...)	Get PyTorch optimizer.
get_time_series_data_loader(dataset[, ...])	Get time series data set data loader.
load_model_state(model[, model_load_state, ...])	Load model state from file.
predict(dataset, model_directory[, model, ...])	Make predictions with recurrent constitutive model for given dataset.
read_best_parameters_from_file(...)	Read best performance state model parameters from file.
read_parameters_history_from_file(...)	Read model learnable parameters history from parameters record file.
save_best_parameters(model, ...)	Save best performance state model parameters.
save_loss_history(model, n_max_epochs, ...)	Save training process loss history record.
save_model_state(model[, state_type, epoch, ...])	Save model state to file.
save_parameters_history(model, ...)	Save model learnable parameters history record.
save_training_state(model, optimizer, state_type)	Save model and optimizer states at given training epoch.
seed_worker(worker_id)	Set workers seed in PyTorch data loaders to preserve reproducibility.
train_model(n_max_epochs, dataset, ...[, ...])	Training of recurrent constitutive model.
write_training_summary_file(device_type, ...)	Write summary data file for model training process.

Classes

EarlyStopper(validation_dataset[, ...])	Early stopping procedure (implicit regularizaton).
RecurrentConstitutiveModel(n_features_in, ...)	Recurrent constitutive model.