edwardchang0112-nctu

How to make custom one (the car plate detection model) to tensorrt?

2

I followed the instructions on the post The steps: ``` python3 save_model.py --weights ./data/custom.weights --output ./checkpoints/custom.tf --input_size 416 --model yolov4 python3 convert_trt.py --weights ./checkpoints/custom.tf --quantize_mode float16 --output ./checkpoints/custom-trt-fp16-416 ``` but...

After running ``` python3 save_model.py --weights ./data/custom.weights --output ./checkpoints/custom-416 --input_size 416 --model yolov4 ``` got the following error messages ``` Non-OK-status: GpuLaunchKernel(FillPhiloxRandomKernelLaunch, num_blocks, block_size, 0, d.stream(), key, counter, gen, data,...

Unable to create group (name already exists) while converting custom weight to yolov4

1

Since I am working on Jetson NX, the tensorflow I used is tensorflow-1.15.4+nv20.12 First, download custom.weight then create a custom.names with only car then create checkpoint directory then just followed...

How can I change the default model when there is no set --model in command line? I want to set default model to be plate detection model without setting --model...

How should I do to input only 1 image instead of video stream, and see the pop up detection results? I already tried ``` python3 detectnet-camera.py --model=./networks/az_plate/az_plate_ssdmobilenetv1.onnx --class_labels=./networks/az_plate/labels.txt --input_blob=input_0 --output_cvg=scores...