SEP-Nets: Small and Effective Pattern Networks · HackerPapers discussions

SEP-Nets: Small and Effective Pattern Networks — mentioned by timanglade

Mon, 26 Jun 2017 16:26:23 GMT

Great question — I did not, because I had unfortunately spent all of my data on that last training run, and I did not have a untainted dataset left to measure the impact of quantization on. (Just poor planning on my part really.) It’s also my understanding at the moment that quantization does not help with inference speed or memory usage, which were my chief concerns. I was comfortable with the binary size (<20MB) that was being shipped and did not feel the need to save a few more MBs there. I was more worried about accuracy, and did not want to ship a quantized version of my network without…

Paper and discussion history. A mention is not an endorsement.

SEP-Nets: Small and Effective Pattern Networks — HN submission

Thu, 15 Jun 2017 01:26:07 GMT

Snapchat paper on how it can put AI on phones

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.

SEP-Nets: Small and Effective Pattern Networks — HN submission

Wed, 14 Jun 2017 03:05:05 GMT

SEP-Nets: Small and Effective Pattern Networks – SNAP Research

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.