kaishijeng

I have converted VOC2012 datasets to KITTI format required by squeezedet. Training is running OK, but converge is very slow if I keep original image size 1248x384 in kitti_squeezeDet_config.py The...

Here is what I did: 1) Follow Training YOLO on VOC in https://pjreddie.com/darknet/yolo to download Pascal VOC dataset. 2) Use my modified voc_pascal_new.py (below) to generate squeezeDet format of label....

See the attachment for z [voc_label_new.zip](https://github.com/BichenWuUCB/squeezeDet/files/1053859/voc_label_new.zip) ip of voc_pascal_new.py

vhtrivedi, Did you have a luck to implement for PascalVOC successfully? I tried it and got an error (See more details in #45 (error in running demo.py). Thanks,

Zifeng, It works with this trick. However, it seems synset is different from the link below because the prediction output is different from actual image. 'http://data.mxnet.io/models/imagenet/resnet/synset.txt'. On Sun, Jan 8,...

I have the same issue. My pytorch is 1.4.0 and it is difficult for me to roll back to 1.1.0 How can use the abn as you suggest below? >>you...

Liu, Are imagenet of solver,prototxt, train_densenet.prototxt and test_densenet.prototxt available? Thanks,

Thanks and will try it

Is imagenet caffe model available yet? Thanks,

@davsol Are you able to make float16 working? Thanks,