Comment by SleekEagle

Exactly, especially in the age of ridiculously rapid development that we have found ourselves in over the past few years. This is exactly why TensorFlow is dying