add new dementions at the front and the tail

sum over the first and last dimention

Execute an replication callback `__data_parallel_replicate__` on each module created by original replication. The callback will be invoked w

Performs a single optimization step. Arguments: closure (callable, optional): A closure that reevaluates the model and returns the l

Args: master_callback: a callback to be invoked after having collected messages from slave devices.

Compute the mean and standard-deviation with sum and square-sum. This method also maintains the moving average on the master device.

Loads TFHub weights and saves them to intermediate HDF5 file. Args: module_path ([Path-like]): Path to TFHub module. hdf5_path ([Path-like]

Checks if a file is an image. Args: filename (string): path to a file Returns: bool: True if the filename ends with a

Numpy implementation of the Frechet Distance. Taken from https://github.com/bioinf-jku/TTUR The Frechet distance between two multivariate Gaussian

Monkey-patch an existing `DataParallel` object. Add the replication callback. Useful when you have customized `DataParallel` implementation.

Register an slave device. Args: identifier: an identifier, usually is the device id. Returns: a `SlavePipe` obj