An inexperienced team is training a neural network model on a desktop computer and this is taking a significant amount of time. What would you recommend to them to speed up model training?
A.
Train the model over multiple desktop computers
B.
Train the model on GPUs
C.
Use a contractor to do the training portion
D.
Break the dataset up into multiple smaller datasets and train the model on each of the smaller datasets over a desktop computer
Training deep neural networks on CPUs is very slow. CPMAI’s Glossary highlights that tensor processing units (TPUs) and GPUs are specialized hardware accelerators explicitly recommended to “accelerate the training and inference of machine learning models” by parallelizing the heavy matrix operations in neural-network layers. Switching from desktop CPU training to GPU-based training can reduce training time by orders of magnitude.
=========
Contribute your Thoughts:
Chosen Answer:
This is a voting comment (?). You can switch to a simple comment. It is better to Upvote an existing comment if you don't have anything to add.
Submit