from Hacker News

Marat_Dukhan

joined 11/3/12, 5:58 AM has 625 karma

• Faster Quantized Neural Network Inference with XNNPack

  by Marat_Dukhan on 9/14/21, 2:35 AM, with 15 comments

• MediaPipe Pose

  by Marat_Dukhan on 5/12/21, 2:14 AM, with 0 comments

• M1 Mac Mini at MacStadium

  by Marat_Dukhan on 12/12/20, 5:30 AM, with 0 comments

• Background Features in Google Meet, Powered by Web ML

  by Marat_Dukhan on 11/2/20, 4:20 AM, with 266 comments

• Supercharging TensorFlow.js with SIMD and multi-threading

  by Marat_Dukhan on 10/26/20, 4:13 AM, with 19 comments

• Introducing the WebAssembly backend for TensorFlow.js

  by Marat_Dukhan on 3/14/20, 4:46 AM, with 0 comments

• Fast Sparse ConvNets

  by Marat_Dukhan on 11/25/19, 11:06 AM, with 1 comments

• Show HN: detecting cache latency inside a Web browser

  by Marat_Dukhan on 1/6/18, 7:52 AM, with 47 comments

• mInference demo: ImageNet convnets inside a Web browser (WAsm/Asm.js/PNaCl)

  by Marat_Dukhan on 5/15/17, 6:55 AM, with 0 comments