A cut & paste of the contents of the PyTorch BatchNorm2d forward function

FBNet-C Paper: https://arxiv.org/abs/1812.03443 Ref Impl: https://github.com/facebookresearch/maskrcnn-benchmark/blob/master/maskrcn

Creates a mnasnet-b1 model. Ref impl: https://github.com/tensorflow/tpu/tree/master/models/official/mnasnet Paper: https://arxiv.org/pdf/1807

Creates a MobileNet-V3 model. Ref impl: ? Paper: https://arxiv.org/abs/1905.02244 Args: channel_multiplier: multiplier to number o

Creates the Single-Path NAS model from search targeted for Pixel1 phone. Paper: https://arxiv.org/abs/1904.02877 Args: channel_multipl

Load weights from .npz checkpoints for official Google Brain Flax implementation

Binary Lovasz hinge loss logits: [B, H, W] Logits at each pixel (between -infinity and +infinity) labels: [B, H, W] Tensor, bi

Multi-class Lovasz-Softmax loss Args: @param probas: [B, C, H, W] Class probabilities at each prediction (between 0 and 1). Int

Adjusts the compatibility of widths and groups.

Create AugMix PyTorch transform :param config_str: String defining configuration of random augmentation. Consists of multiple sections separated

Create a AutoAugment transform :param config_str: String defining configuration of auto augmentation. Consists of multiple sections separate

Returns the names of available CLIP models

Numpy implementation of the Frechet Distance. The Frechet distance between two multivariate Gaussians X_1 ~ N(mu_1, C_1) and X_2 ~ N(mu_2, C_2

feature channels accessor

Calculate score of Chamfer Distance (CD)

Calculate score of Earth Mover's Distance (EMD)

Calculate the score of Fréchet Point-based Volume Distance (FPVD)

Calculate the score of Fréchet Sparse Volume Distance (FSVD)

Calculate the score of Jensen-Shannon Divergence (JSD)

Calculate the score of Minimum Matching Distance (MMD)

Convert applicable model parameters to fp16

cosine schedule as proposed in https://openreview.net/forum?id=-NEXDKk8gZ

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

Generates cropped_image using one of the bboxes randomly distorted. See `tf.image.sample_distorted_bounding_box` for more documentation. Arg