Maarten Bloemen comments

Repositories
Issues
Comments

Results 2 comments of


                                            Maarten Bloemen

Aligning the dataset using GPU does not use GPU, rather just eats up GPU

This is normal [Tensorflow](https://www.tensorflow.org/guide/using_gpu#allowing_gpu_memory_growth) behavior. > By default, TensorFlow maps nearly all of the GPU memory of all GPUs (subject to CUDA_VISIBLE_DEVICES) visible to the process. The low GPU utilisation...

run with Docker

Very nice, great job!