Tensor Cores and deep learning
5 visualizaciones (últimos 30 días)
Mostrar comentarios más antiguos
Gerardo Suarez
el 24 de Jun. de 2019
Comentada: Alexander Sonin
el 3 de Dic. de 2021
Dear all
I have a GeForce RTX 2060. How can I take advantage of its 240 tensor cores for deep learning? I have read that I should set the precision to FP16, but using the "trainnetwork" function there does not seem to be any option.
Thank you
1 comentario
Respuesta aceptada
Krishna Bindumadhavan
el 14 de Sept. de 2019
Editada: Krishna Bindumadhavan
el 14 de Sept. de 2019
Currently, there is no support for training a neural network with half precision data type. This is expected in a future release.
You can however take advantage of tensor cores, while generating code for deploying a trained neural network with GPU Coder. You can look at Deep Learning Prediction with NVIDIA TensorRT example , but set the DataType property of the DeepLearningConfig to 'fp16'. This will use the Tensor cores on a Volta or Turing card such as the RTX series for inference.
In addition, MATLAB supports the half data type via the half precision object in the fixed point designer toolbox: https://www.mathworks.com/help/fixedpoint/ref/half.html?s_tid=doc_ta. General code generation support for the half type, via MATLAB Coder and GPU coder is under active development and can be expected in an upcoming release.
1 comentario
Alexander Sonin
el 3 de Dic. de 2021
Krishna kept silent about the exact answer to this question: MATLAB cannot use tensor cores to train neural networks !!!
Más respuestas (0)
Ver también
Categorías
Más información sobre Deep Learning Toolbox en Help Center y File Exchange.
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!