NVIDIA/tensorrt-laboratory

Issues

• when i have multi place use the TRT in different thread, i hit this crash and caught signal: 11 SIGSEGV

  #39 opened 4 years ago by bigBuffers
  0

• make install does not behave as expected

  #37 opened 4 years ago by ryanolson
  0

• How to write a custom allocator for IGpuAllocator

  #35 opened 4 years ago by ray-lee-94
  1

• TensorRT-7.0 support

  #36 opened 4 years ago by idealboy
  1

• run multiple models at one time on xavier

  #34 opened 5 years ago by jeansely
  1

• Enable gRPC deadlines in Client/Server

  #21 opened 6 years ago by ryanolson
  1

• Creating engines for PyTorch or onnx

  #28 opened 5 years ago by mlcoop
  8

• Build problem

  #30 opened 5 years ago by twmht
  1

• One of your dependencies may have a security vulnerability (Jinja2 < 2.10.1)

  #25 opened 5 years ago by zeroepoch
  1

• Release tags?

  #26 opened 6 years ago by brianthelion
  1

• I found that using tensorrt for inference takes more time than using tensorflow directly on GPU

  #24 opened 6 years ago by jlygit
  1

• Model chaining example

  #22 opened 6 years ago by SlipknotTN
  5

• Back pressure for thread pool task queue

  #19 opened 6 years ago by mrjackbo
  3

• Provide K8s Horizontal Pod AutoScaler Example

  #20 opened 6 years ago by ryanolson
  0

• Extract the TensorRT build as part of the version in the FindTensorRT.cmake

  #17 opened 6 years ago by ryanolson
  1

• Add TensorRT Inference Server Configurations for GTC Demo

  #16 opened 6 years ago by ryanolson
  0

• Post GTC Sides

  #15 opened 6 years ago by ryanolson
  0

• Add IServer Interface

  #14 opened 6 years ago by ryanolson
  0

• DeserializeEngine should accept raw bytes

  #12 opened 6 years ago by ryanolson
  0

• TensorRT::Model::GetBinding should be overloaded to accept input tensor names

  #13 opened 6 years ago by ryanolson
  0

• 03-batching

  #9 opened 6 years ago by mrmeswani
  1

• Improve Metrics and Dashboard

  #7 opened 6 years ago by ryanolson
  0