undefined | Better HN

0 pointsmatt_the_bass7y ago0 comments

My mobile neural nets are still in the training stage. Sometimes, they have trouble recognizing when the dogs fifo is full.

Also, though I derive great happiness from this automation effort, I’ve noticed that the processing units are quite expensive and the maintenance cost is also large. After 5 years, I’m certain I’ve spent more time training them than if I fed the dog myself and more money that if I hired a 3rd party DwaaS (dog walker as a service).

0 comments

delinka7y ago

I think part of the problem is the ordering of the canine calorie queue: should it not be approximately lifo rather than fifo? If you are indeed correct, I’m curious to know how the canine has direct access to the caloric morsels at the bottom of the container.

dudleypippin7y ago

I believe they were referring to the canine unit's output queue, which definitely should be fifo. If it switches to lifo, a trip to the troubleshooting specialist is in order.

matt_the_bassOP7y ago

Yes I was referring the fifo in the state machine that runs throughout the canine unit.

But I do agree that the feeding mechanism should definitely be designed as a lifo

j / k navigate · click thread line to collapse