grimoire/mmdetection-to-tensorrt

Issues

• DDQ-DETR

  #131 opened 3 months ago by mirshah7789
  0

• AttributeError: module 'mmdet2trt.mmdet2trt' has no attribute 'main'

  #129 opened 7 months ago by cudanexus
  3

• Run “mmdet2trt -h”，it shows "ImportError: cannot import name 'create_meshgrid_plugin' from 'torch2trt_dynamic.plugins' (/home/jac/wmj/grid/torch2trt_dynamic/torch2trt_dynamic/plugins/__init__.py)". How can I fix it? I have installed torch2trt_dynamic.

  #130 opened 5 months ago by superprogrammai
  1

• MASKRCNN trt_model has Zero output

  #106 opened 3 years ago by azuryl
  6

• why performance of Mask_rcn tensorrt-fp16_dynamic-320x320-1344x1344 is bad

  #126 opened a year ago by azuryl
  1

• No module named 'torch2trt_dynamic.converters.Conv2d'

  #113 opened 3 years ago by Oswells
  1

• Welcome update to OpenMMLab 2.0

  #125 opened 2 years ago by vansin
  0

• "Myelin Error in MyelinGraphContext" occur while I try to run trt model

  #124 opened 2 years ago by zkccpro
  0

• _pickle.UnpicklingError: unpickling stack underflow

  #123 opened 2 years ago by xuweidongkobe
  0

• Mask(segment) inference error occured.

  #122 opened 2 years ago by jwyng2000
  0

• Thx to your great works. I have some problems on installing Amirstan_plugin. So can you help me?

  #121 opened 2 years ago by jwyng2000
  3

• Converted model.engine does not work properly on DeepStream

  #120 opened 2 years ago by xarauzo
  0

• Converted engine fails on DeepStream: Detect-postprocessor failed to init resource because dlsym failed to get func NvDsInferParseMmdet pointer

  #119 opened 2 years ago by xarauzo
  5

• #assertion /amirstan_plugin/src/plugin/batchedNMSPlugin/batchedNMSPlugin.cpp,127

  #117 opened 2 years ago by Kaeseknacker
  3

• mmdetection maskrcnn to deepstream, 如何支持在deepstream获取mask

  #118 opened 2 years ago by pangchao-git
  1

• Convert DETR to TensorRT Error

  #116 opened 2 years ago by Cverlpeng
  4

• mmdetection 的cascade_dcn_fpn50 转化为tensorrt后，采用本项目，发现转化成deepstream的模型比转化成tensorrt的模型精度较低，有些类阈值很低，有啥排查方法么

  #110 opened 2 years ago by pangchao-git
  2

• deepstream 使用mmdetection-to-tensorrt转换的Mask Rcnn模型 无法检出mask,请问该如何配置.

  #115 opened 2 years ago by pangchao-git
  0

• Unable to convert Side Aware Boundary Localization: [checkSanity.cpp::checkSanity::106] Error Code 2: Internal Error

  #112 opened 3 years ago by MeteoriteWeny
  6

• Unable to convert FRCNN: Warning: Encountered known unsupported method torch.Tensor.new_tensor

  #111 opened 3 years ago by Kaeseknacker
  3

• unsupported operand type(s)

  #109 opened 3 years ago by nvvnst
  1

• new mmdetection env converted trt engine inference error

  #107 opened 3 years ago by lucasjinreal
  3

• Unable to generate mask results.

  #85 opened 3 years ago by tehkillerbee
  8

• MASKRCNN Encountered known unsupported method torch.Tensor.new_tensor

  #105 opened 3 years ago by azuryl
  7

• Can not load dataset from config. Use default CLASSES instead.

  #104 opened 3 years ago by azuryl
  3

• segfault while yolox engine making on TensorRT 8

  #98 opened 3 years ago by victor-yudin
  3

• do you have a plan for windows

  #76 opened 3 years ago by lqyp0420
  8

• Size Mismatch

  #102 opened 3 years ago by selimceylan
  2

• Python logging should be defined for the module name - not root logger.

  #95 opened 3 years ago by tehkillerbee
  1

• inference score become low

  #99 opened 3 years ago by Chen-cyw
  0

• Setting opt_shape_param min_shape==opt_shape==max_shape does not work correctly.

  #97 opened 3 years ago by tehkillerbee
  2

• Does it handles fuse conv and bn when convert to trt?

  #93 opened 3 years ago by lucasjinreal
  1

• Unable to convert HTC with DCN and multi-scale training model to tensor-rt

  #94 opened 3 years ago by Prasoonk02
  1

• how to load generated model back in python?

  #91 opened 3 years ago by lucasjinreal
  7

• Can not convert custom mmdet to tensorrt

  #92 opened 3 years ago by selimceylan
  2

• FCOS head Bug：x1 y1 will be overstep the boundary and are set to 0。

  #88 opened 3 years ago by Xingxu1996
  1

• Wrong pytorch-cuda version installation with pip

  #82 opened 3 years ago by mmeendez8
  2

• Deformable DETR

  #81 opened 3 years ago by fduxvtech
  3

• inference cascade_rcnn has error

  #77 opened 3 years ago by lqyp0420
  0

• Wrong detection boxes with FCOS detector

  #75 opened 3 years ago by guillaume-michel
  8

• Problem with fp16 detection results (zero x value)

  #74 opened 3 years ago by Kaeseknacker
  5

• Does model.engine depend on CUDA version?

  #70 opened 3 years ago by Kaeseknacker
  3

• Before and after the conversion，the accuray between is not stable.

  #73 opened 3 years ago by caiduoduo12138
  1

• TypeError: __init__() got an unexpected keyword argument 'init_cfg'

  #72 opened 3 years ago by xizi
  1

• Unable to use converted YOLOv3 model

  #71 opened 3 years ago by m-nny
  5

• Unable to fetch in submodule path 'third_party/cub'

  #68 opened 3 years ago by BChunlei
  2

• solov2 support

  #67 opened 3 years ago by elepherai
  1

• Can't open output tensorrt model with Netron

  #66 opened 3 years ago by chienkan
  1

• what is the difference between mmdetection-to-tensorrt and torch2trt

  #65 opened 3 years ago by liuhuiCNN
  1

• ValueError: __len__() should return >= 0

  #63 opened 3 years ago by Kaeseknacker
  4