API and modules
Function
|
Run data integration |
|
Reweight labels to make all cell types share the same total weight |
|
Load single cell dataset from file |
|
Filter cells and genes |
|
Batch-specific scale data |
DataLoader
|
|
|
Load data for training. |
Model
|
Variational Autoencoder framework |
|
Domain-specific Batch Normalization |
|
Basic block consist of: |
|
Neural network consist of multi Blocks |
|
VAE Encoder |
|
VAE Decoder |
|
|
|
Returns the matrix of ||x_i-y_j||_p^p. |
|
Calculate a Wasserstein distance matrix between the gmm distributions with diagonal variances |
|
Calculate a unbalanced optimal transport matrix between mini batches. |
|
Make the input tensor one hot tensors |
|
Early stops the training if loss doesn't improve after a given patience. |
Evaluation
|
Calculate batch entropy mixing score |
|
Wrapper for sklearn silhouette function values range from [-1, 1] with |
|
Label transfer |
Logger
|