virne.solver.learning.utils

Functions

apply_mask_to_logit(logit[, mask])	Apply a mask to a given logits tensor.
get_available_device()	Get the available device (CPU or GPU).
get_observations_sample(obs_batch, indices)	Get a sample from an input observation batch given the indices.
get_pyg_batch(x_batch, edge_index_batch[, ...])	Convert a batch of node and edge information into Pytorch Geometric format.
get_pyg_data(x, edge_index[, edge_attr])	Convert node and edge information into Pytorch Geometric format.
load_pyg_batch_from_network_list(network_list)	Load batch data from a list of networks
load_pyg_data_from_network(network[, ...])	Load data from network
normailize_data(data[, method])	Normalize node or edge data.
update_mean_var_count_from_moments(mean, ...)	Update mean, variance, and count from the batch mean, variance, and count.

Classes

RunningMeanStd([epsilon, shape])

Calculate running mean and standard deviation for a given data.