Paul Irish
26.9K
posts
Been loving NVIDIA's NeMo models recently to complement Whisper for transcription, but the startup felt a bit heavy. So I went on a profiling adventure! github.com/NVIDIA-NeMo/Ne… Turned out to be fruitful… stoked to land a big optimization in NeMO. github.com/NVIDIA-NeMo/Ne…
- @flumewater
I am absolutely loving the accuracy and data granularity. Not only have I verified it's only 0.083% off the neptune t10 meter.. but I can see flow rate differences for each station of the sprinkler system! Time to cut these numbers down. :) Thank you!

