• GPU machin : mlwyberns.sogang.ac.kr

• 컨테이너 생성/실행

  • 사용된 이미지 : musing_darwin_backup

  nvidia-docker run -i -t --name [컨테이너 이름] musing_darwin_backup /bin/bash
  

• 데이터 파일 놓을 위치

  /root/tf-faster-rcnn/data
  

1. Create a folder and a soft link to use the pre-trained model

NET=res101
TRAIN_IMDB=voc_2007_trainval+voc_2012_trainval
mkdir -p output/${NET}/${TRAIN_IMDB}
cd output/${NET}/${TRAIN_IMDB}
ln -s ../../../data/voc_2007_trainval ./default
cd ../../..

2. Download pre-trained models and weights.

   For Resnet101, you can set up like:

   mkdir -p data/imagenet_weights
   cd data/imagenet_weights
   wget -v http://download.tensorflow.org/models/resnet_v1_101_2016_08_28.tar.gz
   tar -xzvf resnet_v1_101_2016_08_28.tar.gz
   mv resnet_v1_101.ckpt res101.ckpt
   cd ../..

3. Train (and test, evaluation)

./experiments/scripts/train_faster_rcnn.sh [GPU_ID] [DATASET] [NET]
# GPU_ID is the GPU you want to test on
# NET in {vgg16, res50, res101, res152} is the network arch to use
# DATASET {pascal_voc, pascal_voc_0712, coco} is defined in train_faster_rcnn.sh
# Examples:
./experiments/scripts/train_faster_rcnn.sh 1 pascal_voc res101
./experiments/scripts/test_faster_rcnn.sh 0 pascal_voc_0712 res101

Note: Please double check you have deleted soft link to the pre-trained models before training. If you find NaNs during training, please refer to Issue 86. Also if you want to have multi-gpu support, check out Issue 121.

4. Visualization with Tensorboard

tensorboard --logdir=tensorboard/vgg16/voc_2007_trainval/ --port=7001 &
tensorboard --logdir=tensorboard/vgg16/coco_2014_train+coco_2014_valminusminival/ --port=7002 &

5. Test and evaluate

./experiments/scripts/test_faster_rcnn.sh [GPU_ID] [DATASET] [NET]
# GPU_ID is the GPU you want to test on
# NET in {vgg16, res50, res101, res152} is the network arch to use
# DATASET {pascal_voc, pascal_voc_0712, coco} is defined in test_faster_rcnn.sh
# Examples:
./experiments/scripts/test_faster_rcnn.sh 0 pascal_voc res101
./experiments/scripts/test_faster_rcnn.sh 1 coco res101

6. You can use tools/reval.sh for re-evaluation

By default, trained networks are saved under:

output/[NET]/[DATASET]/default/

Test outputs are saved under:

output/[NET]/[DATASET]/default/[SNAPSHOT]/

Tensorboard information for train and validation is saved under:

tensorboard/[NET]/[DATASET]/default/
tensorboard/[NET]/[DATASET]/default_val/

The default number of training iterations is kept the same to the original faster RCNN for VOC 2007, however I find it is beneficial to train longer (see report for COCO), probably due to the fact that the image batch size is one. For VOC 07+12 we switch to a 80k/110k schedule following R-FCN. Also note that due to the nondeterministic nature of the current implementation, the performance can vary a bit, but in general it should be within ~1% of the reported numbers for VOC, and ~0.2% of the reported numbers for COCO. Suggestions/Contributions are welcome.