search

chenyuntc / simple-faster-rcnn-pytorch

A simplified implemention of Faster R-CNN that replicate performance from origin paper
Other
3.98k stars 1.14k forks source link

How to run by pretrained model from torchvision? #66

Open shangjianan2 opened 6 years ago

shangjianan2 commented 6 years ago

How to run by pretrained model from torchvision? Just skip "5.2 Prepare caffe-pretrained vgg16" and run “mkdir checkpoints/” and "python3 train.py train --env='fasterrcnn-caffe' --plot-every=100 --caffe-pretrain"? Or run "python3 train.py train" instead of "python3 train.py train --env='fasterrcnn-caffe' --plot-every=100 --caffe-pretrain"? ps: because I find "caffe_pretrain" in "config.py" is false which I guess the code will run by pretrained model from torchvision if I run "python3 train.py train". Am I right?

shangjianan2 commented 6 years ago

If I run "python3 train.py train --env='fasterrcnn-caffe' --plot-every=100 --caffe-pretrain", the "caffe_pretrain" in "config.py" is true. But I want to run by pretrained model from torchvision for I can not download pretrained model from caffe.

shangjianan2 commented 6 years ago

I run “python3 train.py train --test-num=1”. The output is "1264it [04:04, 5.17it/s]'>' not supported between instances of 'float' and 'NoneType'". Why there is "'>' not supported between instances of 'float' and 'NoneType'"?