Читать книгу Machine Learning for Tomographic Imaging - Professor Ge Wang - Страница 56

Poisson

Оглавление

The Poisson loss function offers a measure on how the predicted distribution diverges from the expected distribution. The Poisson distribution is widely used for modeling photon-count data, which is highly relevant to medical imaging signals. The Poisson loss function is computed as follows:

L=1n∑i=1nyˆ(i)−y(i)·logyˆ(i).(3.23)

Machine Learning for Tomographic Imaging

Подняться наверх