pytorch/QNNPACK

Issues

• Qnnpack accuracy very poor on unet model

  #73 opened 4 years ago by Amitdedhia6
  3

• Support for nn.ConvTranspose2d

  #72 opened 4 years ago by Amitdedhia6
  0

• How to create a quantized model in order to try out QNNPACK?

  #12 opened 6 years ago by fengxueem
  13

• Understanding XZP ukernel

  #70 opened 5 years ago by wenfahu
  1

• Build failed on rasp pi 4 (Error: selected processor does not support `yield' in ARM mode)

  #71 opened 4 years ago by pyjhzwh
  1

• Is QNNPACK still being developed?

  #69 opened 5 years ago by gibiansky
  1

• How to build dependencies separately

  #68 opened 5 years ago by adamjstewart
  0

• COMPAIR ERROR

  #67 opened 5 years ago by mathpopo
  2

• build error

  #61 opened 5 years ago by lucasjinreal
  0

• Is the code multithreaded for arm64?

  #60 opened 5 years ago by RuABraun
  7

• is there any example to use fp16 gemm？

  #62 opened 5 years ago by HaoLiuHust
  0

• 3d convolution support

  #59 opened 5 years ago by dmenig
  1

• Will you going to open source QNNPACK quantize tool and methods?

  #43 opened 6 years ago by csuhawk
  2

• Typo in the announcement blog?

  #55 opened 6 years ago by zhenhuaw-me
  1

• Crashed when running conv3x3 test

  #53 opened 6 years ago by wonderzy
  0

• Request to add a Concat layer

  #52 opened 6 years ago by E1eMenta
  1

• What's the relationship between the ASM and C version micro kernel?

  #50 opened 6 years ago by zhenhuaw-me
  5

• Deconvolutional Layer is slow for large kernel size

  #46 opened 6 years ago by yangwenca
  2

• Request to add Conv1d related operation

  #48 opened 6 years ago by viggi1000
  1

• Semantic segmentation examples

  #37 opened 6 years ago by brucechou1983
  1

• Could you pls provide Android Apk demo?

  #32 opened 6 years ago by Meicsu199345
  6

• Implementation detail things...

  #45 opened 6 years ago by zhenhuaw-me
  9

• No performance difference between ARMv7 (32-bit) Android and ARM64 (64-bit) Android

  #47 opened 6 years ago by yangwenca
  1

• No documentation? or how to execute a layer in QNNPACK

  #42 opened 6 years ago by RuABraun
  8

• Per channel quantization for the weights

  #44 opened 6 years ago by RuABraun
  1

• do you have some detail resource about quantize ?

  #3 opened 6 years ago by victorygogogo
  3

• Interested in using QNNPACK operators as part of another C++ library

  #41 opened 6 years ago by RuABraun
  2

• Weird performance gap observed for Jetson TX2

  #29 opened 6 years ago by TerryTsao
  7

• How can I create a quantized model by QNNPACK?

  #35 opened 6 years ago by Meicsu199345
  2

• Can you provide resnet_50_quant example?

  #38 opened 6 years ago by mosjoker
  1

• multi-thread problem.

  #36 opened 6 years ago by 32l
  7

• how can I use qnnpack to recognize a cat?

  #34 opened 6 years ago by dvlee1024
  2

• 8bit depthwise convolution is much slow than tf-lite 8bit!!!

  #33 opened 6 years ago by 32l
  2

• caffe2 abort when running benchmark with input_type uint8_t

  #31 opened 6 years ago by zhenhuaw-me
  4

• Out-of-memory during compiling on Raspbian

  #28 opened 6 years ago by sheh
  4

• Why QNNPACK has better accuracy on MobileNetV2?

  #14 opened 6 years ago by Lyken17
  4

• qnnpack/src/init.c:0: error: undefined reference to 'q8dw_ukernel_25c8__neon' clang++: error: linker command failed with exit code 1 (use -v to see invocation)

  #22 opened 6 years ago by jianlongf
  0

• Error compiling on PC

  #17 opened 6 years ago by orentadmor
  2

• Can I convert a PyTorch model in float to qnnpack model

  #10 opened 6 years ago by ryanqiutu
  4

• Performance benchmark on GPUs, NPUs

  #11 opened 6 years ago by chaos5958
  1

• How to use QNNPACK on iOS?

  #9 opened 6 years ago by hcsearch
  1

• End to end benchmark results

  #1 opened 6 years ago by llhe
  2

• Building on OSX 10.14 for OS64 (ARM 64) fails

  #4 opened 6 years ago by kohenchia
  4

• Building on OSX 10.14 for SIMULATOR64 (i386) fails

  #5 opened 6 years ago by kohenchia
  2

• How use the QNNPACK in PyTorch1.0 ?

  #6 opened 6 years ago by Aspirinkb
  2

• Handler of variate in creating convolution23_nhwc_q8 function

  #7 opened 6 years ago by JSnobody
  1

• build-local.sh error!

  #2 opened 6 years ago by clhne
  5

• Javascript version of the Library

  #8 opened 6 years ago
  0