seongkyun

To short, 1. update the resnet50.onnx file by running 'python3 ./pytorch_model.py' in the terminal. 2. change the function named "parseOnnxModel" from `TRTUniquePtr network{builder->createNetwork()};` to ` const auto explicitBatch = 1U...

Because that line occurs an error Building the model... Traceback (most recent call last): File "main.py", line 69, in main() File "main.py", line 35, in main model = ShuffleNet(config_args) File...

check the `coco.py` file.

English please :)

That is exactly the same 3x3 convolution. (3x1 conv + 1x3 conv = 3x3 conv) The reason for doing this is just for the calculation efficiency. (Very few accuracy dropping...

Maybe, I think that firstly, you should read the Single Shot Multi-box Detector paper (https://arxiv.org/abs/1512.02325).

Multibox loss can be applied to any kind of single shot detectors. That is a kinda loss function.

Exactly yes. You know that there is any code that can train Yolo V3 with any other deep learning toolkit except Darknet. Every other codes are just reads pre-trained weight...

You can download that file on here. https://dl.dropboxusercontent.com/s/s3tw5zcg7395368/instances_valminusminival2014.json.zip After download, just change file's name correctly(instances_trainval35k.json) and copy that to correct directory(annotations/). Then, you can train with MS COCO dataset.