Korabelnikov Aleks comments

Results 12 comments of


                                            Korabelnikov Aleks

the inference speed is much slower than original TensorFlow code

Maybe this implementation is wrong and/or there are some memory transfers across devices?

the inference speed is much slower than original TensorFlow code

but why is the difference when both torch and tf use cudnn?

TensorRT for EfficientNet model

> Hey! I'm trying to convert the model to TensorRT (ver. 5.1.6.1) and surprisingly everythin did you resolve it?

export_graph takes long time and no output file

Thank you for a fast response 2020-12-09 12:39:55.209331: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library cudart64_101.dll Python version: 3.8.5 (default, Sep 3 2020, 21:29:08) [MSC v.1916 64 bit (AMD64)] ONNX version:...

export_graph takes long time and no output file

[faster_rcnn_mobilenetv2_backbone_only_224x224.zip](https://github.com/onnx/onnx-tensorflow/files/5675712/faster_rcnn_mobilenetv2_backbone_only_224x224.zip) It seems issue occurs under Windows only

demo not accurate

@twangnh and @FrancescoFornasa I have exactly same result. Have you overcome it? Have you met difficulties when retrained?

"Object was never used" errors after upgrading to Tensorflow 1.2

me fifteen. Any suggestions?

Other models available other than 4-stacked?

Any changes in subject? Does not-stacked model available now?

Delays when transcribing streaming audio

@scott-vector Hey, have you managed to fix delays?

Ubuntu14.04 server x64, OpenCV3, CuDNNv3, make errors.

It caused by inconsistent caffe and openCV. I have met such error. Just remove line with redefinion