TensorFlow: How to optimise your input pipeline with queues and multi-threading
Morgan
29312

You should definitely check out the awesome library called “TensorPack” which does the prefetching (written in python) in an advanced way using ZMQ and multiple processed to avoid GIL.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.