jjldr

How to set config.py and prior_box.py when input images are not resized to 300*300 or 512*512

2

When I use RFBNet for KITTI 2D detection, I struggle with input image size. The image size of KITTI is 375 * 1242, the aspect ratio is almost 1:3, so...

You can find c++ implementation of another method in reference[2].

Can you provide loss curve during training

10

When training 3D object detection on my dataset, the rot_loss is much larger than all other losses, is it normal ? Below is part of my training loss: ...

Cann't load pretrained .weights file

8

Thanks for you share.however,I cann't load you pretrained weights.

Thank you for your sharing. However, the train code is not included, please upload your training code, thx.

can not convert from .t7

4

Thank you for sharing your code. But when i run the t7_to_pth.py with PyTorch 0.4.1, some error happens,the error is as follows: File "t7_to_pth.py", line 93, in model2[prefix+'conv_d.weight'] = model1.modules[42].modules[6].modules[0].modules[0].modules[2].modules[0].modules[1].modules[1].weight...