Semantic segmentation of human brain cortex with UNET-like architecture using SONY Neural Network Console

Training was performed using NVIDIA Titan RTX video card with 24 GB of memory. With presented dataset of ~2700 RGB images, each 320 by 320 pixels, training speed was 2 minutes per one epoch. It has to be outlined that NNABLA library is using CUDA and cuDNN architecture very efficiently, with GPU utilization rate reaching 90%-95%.

Below is the example of the learning curve:


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.