1ytic/warp-rnnt

Issues

• Exception: CPU version is not implemented

  #37 opened 2 years ago by iflov
  2

• __version__ assignment breaks local build.

  #41 opened 7 months ago by JackTemaki
  0

• ImportError: libcudart.so.10.1: cannot open shared object file: No such file or directory

  #40 opened 10 months ago by hechengzhi985
  2

• undefined symbol: _ZNSt19basic_ostringstreamIcSt11char_traitsIcESaIcEEC1Ev

  #33 opened 3 years ago by wylqq312715289
  2

• rnnt_loss status 1

  #34 opened 3 years ago by jiay7
  2

• can't install warp-rnnt

  #36 opened 2 years ago by 12-zhx
  2

• runtime error with Python 3.8

  #4 opened 4 years ago by funcwj
  4

• improve efficiency of warps

  #38 opened 2 years ago by maxwellzh
  4

• THC/THC.h: No such file or directory

  #35 opened 2 years ago by cchan-lm
  7

• Strange behavior using PyTorch DDP

  #32 opened 3 years ago by snakers4
  7

• WARNING: sample 0 [42, 26] has a forward/backward mismatch -52.543503 / 0.000000

  #19 opened 4 years ago by songtaoshi
  3

• question about the rnnt loss arguments

  #18 opened 4 years ago by songtaoshi
  4

• RuntimeError: rnnt_loss status 1

  #22 opened 3 years ago by CLC2019
  5

• PyTorch 1.9 Support

  #31 opened 3 years ago by snakers4
  5

• Normalize the RNN-T Loss with input seq length

  #30 opened 3 years ago by aheba
  1

• Not support for pytorch 1.7

  #28 opened 3 years ago by Jxu-Thu
  1

• ninja: build stopped: subcommand failed.

  #29 opened 3 years ago by harisgulzar1
  1

• ImportError: libcudart.so.10.2: cannot open shared object file: No such file or directory

  #20 opened 3 years ago by skgusrb12
  3

• Question about average_frames and reduction parmas

  #27 opened 3 years ago by wl-junlin
  1

• question about the gather arguments

  #25 opened 3 years ago by wl-junlin
  1

• Benchmarking methodology used is not quite correct

  #9 opened 3 years ago by AshishSardana
  6

• warning that forward/backward mismatch

  #24 opened 3 years ago by maxwellzh
  3

• Transducer loss leads to memory leak

  #21 opened 3 years ago by YangHao97
  4

• RuntimeError: Trying to backward through the graph a second time, but the buffers have already been freed。

  #12 opened 4 years ago by lfgogogo
  1

• Exception: CPU version is not implemented

  #17 opened 4 years ago by yyMoming
  1

• Undefined symbol when loading module

  #13 opened 4 years ago by mjurkus
  5

• Can't install

  #11 opened 4 years ago by jeongHwarr
  3

• how to apply backward to the output of rnnt_loss

  #10 opened 4 years ago by huangnengCSU
  0

• Problem installation

  #5 opened 4 years ago by roholazandie
  3

• Not support Pytorch 1.5

  #8 opened 4 years ago by li563042811
  1

• Not support pytorch 1.6

  #7 opened 4 years ago by wyattsuen
  1

• Issues with non-standard CUDA install

  #6 opened 4 years ago by StarDuster
  4

• Not compatible with latest pytorch like 1.5.0

  #1 opened 5 years ago by jacob1017
  4

• operating with apex?

  #3 opened 5 years ago by tongjinle123
  1