coco_vgg16_faster_rcnn_final.caffemodel or VGG16.v2.caffemodel

[monajalal]

Should I use coco_vgg16_faster_rcnn_final.caffemodel from: wget www.cs.berkeley.edu/~rbg/faster-rcnn-data/coco_vgg16_faster_rcnn_final.caffemodel or should I use the .caffemodel in data/imagenet_models?

mona@pascal:~/computer_vision/py-faster-rcnn/data/imagenet_models$ ls
VGG16.v2.caffemodel  VGG_CNN_M_1024.v2.caffemodel  ZF.v2.caffemodel

What's the difference and should I stick to the one that is already in imagenet_models for the following command?

tools/train_net.py \
    --gpu 0 \
    --solver ./models/coco/VGG16/faster_rcnn_end2end/solver.prototxt \
    --weights data/imagenet_models/VGG16.v2.caffemodel \
    --imdb coco_2014_train+coco_2014_valminusminival \
    --iters 490000 \
    --cfg ./experiments/cfgs/faster_rcnn_end2end.yml

[jtdutta1]

Hi. Can you tell me what you did?