Move the batch to device Args: device (torch.device): the device to move to. non_blocking (bool, optional): whether t

Split the batch among dim=0 into chunks. The meta_info is passed to each DataProto after split. Args: split_size (int): the size

Pop a subset of the DataProto via `batch_keys` and `meta_info_keys` Args: batch_keys (list, optional): a list of strings indicati

Get value by key at specified index (indices) if idx == None, returns value for key at each output index if idx is an integer, return

Create an optimizer. TODO currently the model is passed in and all parameters are selected for optimization. For more general use an interfa

r"""Fills the input Tensor with values drawn from a truncated normal distribution. The values are effectively drawn from the normal distributi

Build model with specified default_cfg and optional model_cfg This helper fn aids in the construction of a model including: * handling def

Select a 2d convolution implementation based on arguments Creates and returns one of torch.nn.Conv2d, Conv2dSame, MixedConv2d, or CondConv2d.

r"""A helper function for checkpointing sequential models. Sequential models execute a list of modules/functions in order (sequentially). The

Feature map down-sampling.

Update kl_coef according to current KL.

Activation Layer Factory Fetching activation layers by name with this function allows export or torch script friendly functions to be returne

Creates an EfficientNet model. Ref impl: https://github.com/tensorflow/tpu/blob/master/models/official/efficientnet/efficientnet_model.py Pap

interleave 2 block types in stack

Repeat the batch data a specified number of times. Args: repeat_times (int): Number of times to repeat the data.

Decode block architecture definition strings -> block kwargs Args: arch_def: architecture definition strings, list of list of strings

Create a model. Lookup model's entrypoint function and pass relevant args to create a new model. <Tip> **kwargs will be passed throu

Feature map interpolation.

generate N-D grid in dimension order. The ndgrid function is like meshgrid except that the order of the first two input arguments are switched.

Split the batch among dim=0 into chunks. The meta_info is passed to each DataProto after split. Args: chunks (int): the number of

Performs a single optimization step.

This method forwards all its arguments to SiglipTokenizer's [`~PreTrainedTokenizer.decode`]. Please refer to the docstring of this me