Bichen Wu

Please check-out the newest commit. Now demo.py can also load SqueezeDet+. But you need to use the flag `--demo_net=squeezeDet+` and `--checkpoint=./data/model_checkpoints/squeezeDetPlus/model.ckpt-95000` together so it can load the right model specification...

Hi Sorry for the delay in replying. Thanks for letting me know. I'll take a look into that and integrate the updated evaluation script. Yes, I changed image resolution (1242x375...

@ByeonghakYim Thanks for your question. I'm also curious to see how the IOU-thresholding based box matching works for you. Could you try to explain a bit more on what you...

@ByeonghakYim > so that each ground truth has one anchor. but I could find that some of them has under 0.1 top IOU I wonder what is the reason why...

An anchor is not going to be matched with multiple ground truth boxes. At this [line](https://github.com/BichenWuUCB/squeezeDet/blob/master/src/dataset/imdb.py#L201) and below, you can see how this is handled.

@ShervinAr Thanks for your question. Currently, I don't have a plan to re-implement SqueezeDet in Caffe or other deep learning tools. But if you only need inference or deployment, it...

You can find the model definition from [here](https://github.com/BichenWuUCB/squeezeDet/blob/master/src/nets/squeezeDet.py)

^src/eval.py is calling KITTI's evaluation script. Have you compiled it before running evaluation?