Data Reader for Cityscapes Dataset https://www.cityscapes-dataset.com Data is derived from CityScapes, and can be downloade

Parallel Bottleneck Channel-Spatial Attention Block

Bi-directional Feature Pyramid Network

Bi-directional Feature Pyramid Network

Lighting noise(AlexNet - style PCA - based noise)

Activated Batch Normalization This gathers a `BatchNorm2d` and an activation function in a single module

Implements AdaBound algorithm. It has been proposed in `Adaptive Gradient Methods with Dynamic Bound of Learning Rate`_. Arguments: pa

Computes and stores the average and current value

Channel Attention Block

DSN : We need to consider two supervision for the model.

Calculate loss in multiple-GPUs, which balance the memory usage for Semantic Segmentation. The targets are splitted across the specified

Implements data parallelism at the module level. This container parallelizes the application of the given module by splitting the input acros

ConvNet block for building DenseASPP.

r"""Randomly zeroes spatial blocks of the input tensor. As described in the paper `DropBlock: A regularization method for convolutional networ

Focal Loss, which is proposed in: "Focal Loss for Dense Object Detection (https://arxiv.org/abs/1708.02002v2)"

InPlace Activated Batch Normalization

InPlace Activated Batch Normalization with cross-GPU synchronization This assumes that it will be replicated across GPUs using the same mechanism

Learning Rate Scheduler Step mode: ``lr = baselr * 0.1 ^ {floor(epoch-1 / lr_step)}`` Cosine mode: ``lr = baselr * 0.5 * (1 + cos

Position Attention Block