A 36% price cut to GPU instances, in addition to the potential new benefits offered by software and GPU updates, however, might be enough to tip the cost-efficiency scales back in favor of GPUs.
Thanks to Keras, performing deep learning on a very large number of Reddit submissions is actually pretty easy. Performing it well is a different story.
Keras + TensorFlow + Pretrained character embeddings makes text generation a breeze.
Given that a SF police arrest occurs at a specified time and place, what is the reason for that arrest?
Fancy machine learning approaches may not be required to help Redditors discover new things.