Name of the module/class e.g. torch.nn.functional.

Copy model gradients to master gradients. Args: model_params: List of model parameters created by :func:`prep_param_lists`.

Convert model to half precision in a batchnorm-safe way. Retained for legacy purposes. It is recommended to use FP16Model.

compute cosine distance between two martrix x and y with sizes (n1, d), (n2, d)

compute eculidean distance between two martrix x and y with sizes (n1, d), (n2, d)

Returns the number of training cameras.

Returns the number of training person identities.

Name of the operator e.g. sigmoid.

r"""Opens all layers in model for training. Examples:: >>> from lreid.utils import open_all_layers >>> open_all_layers(model)

r"""Opens specified layers in model for training while keeping other layers frozen. Args: model (nn.Module): neural net model.

r"""Opens specified layers in model for training while keeping other layers frozen. Args: model (nn.Module): neural net m