Max-Pooling Loss

Loss Max-Pooling for Semantic Image Segmentation

Installation

Requirements

To install PyTorch, please refer to https://github.com/pytorch/pytorch#installation.

Compiling

Some parts of Max-Pooling Loss have a native C++ implementation, which must be compiled with the following commands:

cd mpl
python build.py

Using

import mpl
import torch

max_pooling_loss = mpl.MaxPoolingLoss(ratio=0.3, p=1.7, reduce=True)
loss = torch.Tensor(1, 3, 3, 3).uniform_(0, 1)
loss = max_pooling_loss(loss)