Faster-rcnn.pytorch: train with coco2014

[Markovcom]

CUDA_VISIBLE_DEVICES=1 python trainval_net.py --dataset coco -- --bs 3 --nw 5 --lr 0.001 --lr_decay_step 5 --cuda
Called with args:
Namespace(batch_size=3, checkepoch=1, checkpoint=0, checkpoint_interval=10000, checksession=1, class_agnostic=False, cuda=True, dataset='coco', disp_intee=False, lr=0.001, lr_decay_gamma=0.1, lr_decay_step=5, mGPUs=False, max_epochs=20, net='res101', num_workers=5, optimizer='sgd', resume=False, save_dir=sion=1, start_epoch=1, use_tfboard=False)
Using config:
{'ANCHOR_RATIOS': [0.5, 1, 2],
'ANCHOR_SCALES': [4, 8, 16, 32],
'CROP_RESIZE_WITH_MAX_POOL': False,
'CUDA': False,
'DATA_DIR': '/home/code/zhoulinmao/faster-rcnn.pytorch/data',
'DEDUP_BOXES': 0.0625,
'EPS': 1e-14,
'EXP_DIR': 'res101',
'FEAT_STRIDE': [16],
'GPU_ID': 0,
'MATLAB': 'matlab',
'MAX_NUM_GT_BOXES': 50,
'MOBILENET': {'DEPTH_MULTIPLIER': 1.0,
'FIXED_LAYERS': 5,
'REGU_DEPTH': False,
'WEIGHT_DECAY': 4e-05},
'PIXEL_MEANS': array([[[ 102.9801, 115.9465, 122.7717]]]),
'POOLING_MODE': 'align',
'POOLING_SIZE': 7,
'RESNET': {'FIXED_BLOCKS': 1, 'MAX_POOL': False},
'RNG_SEED': 3,
'ROOT_DIR': '/home/code/zhoulinmao/faster-rcnn.pytorch',
'TEST': {'BBOX_REG': True,
'HAS_RPN': True,
'MAX_SIZE': 1000,
'MODE': 'nms',
'NMS': 0.3,
'PROPOSAL_METHOD': 'gt',
'RPN_MIN_SIZE': 16,
'RPN_NMS_THRESH': 0.7,
'RPN_POST_NMS_TOP_N': 300,
'RPN_PRE_NMS_TOP_N': 6000,
'RPN_TOP_N': 5000,
'SCALES': [600],
'SVM': False},
'TRAIN': {'ASPECT_GROUPING': False,
'BATCH_SIZE': 128,
'BBOX_INSIDE_WEIGHTS': [1.0, 1.0, 1.0, 1.0],
'BBOX_NORMALIZE_MEANS': [0.0, 0.0, 0.0, 0.0],
'BBOX_NORMALIZE_STDS': [0.1, 0.1, 0.2, 0.2],
'BBOX_NORMALIZE_TARGETS': True,
'BBOX_NORMALIZE_TARGETS_PRECOMPUTED': True,
'BBOX_REG': True,
'BBOX_THRESH': 0.5,
'BG_THRESH_HI': 0.5,
'BG_THRESH_LO': 0.0,
'BIAS_DECAY': False,
'BN_TRAIN': False,
'DISPLAY': 20,
'DOUBLE_BIAS': False,
'FG_FRACTION': 0.25,
'FG_THRESH': 0.5,
'GAMMA': 0.1,
'HAS_RPN': True,
'IMS_PER_BATCH': 1,
'LEARNING_RATE': 0.001,
'MAX_SIZE': 1000,
'MOMENTUM': 0.9,
'PROPOSAL_METHOD': 'gt',
'RPN_BATCHSIZE': 256,
'RPN_BBOX_INSIDE_WEIGHTS': [1.0, 1.0, 1.0, 1.0],
'RPN_CLOBBER_POSITIVES': False,
'RPN_FG_FRACTION': 0.5,
'RPN_MIN_SIZE': 8,
'RPN_NEGATIVE_OVERLAP': 0.3,
'RPN_NMS_THRESH': 0.7,
'RPN_POSITIVE_OVERLAP': 0.7,
'RPN_POSITIVE_WEIGHT': -1.0,
'RPN_POST_NMS_TOP_N': 2000,
'RPN_PRE_NMS_TOP_N': 12000,
'SCALES': [600],
'SNAPSHOT_ITERS': 5000,
'SNAPSHOT_KEPT': 3,
'SNAPSHOT_PREFIX': 'res101_faster_rcnn',
'STEPSIZE': [30000],
'SUMMARY_INTERVAL': 180,
'TRIM_HEIGHT': 600,
'TRIM_WIDTH': 600,
'TRUNCATED': False,
'USE_ALL_GT': True,
'USE_FLIPPED': True,
'USE_GT': False,
'WEIGHT_DECAY': 0.0001},
'USE_GPU_NMS': True}
zlm: train
loading annotations into memory...
Done (t=18.82s)
creating index...
index created!
Loaded dataset coco_2014_train for training
Set proposal method: gt
Appending horizontally-flipped training examples...
coco_2014_train gt roidb loaded from /home/code/zhoulinmao/faster-rcnn.pytorch/data/cache/coco_2014_train_gt_roidb.pkl
done
Preparing training data...
done
zlm: valminusminival
loading annotations into memory...
Traceback (most recent call last):
File "trainval_net.py", line 202, in
imdb, roidb, ratio_list, ratio_index = combined_roidb(args.imdb_name)
File "/home/code/zhoulinmao/faster-rcnn.pytorch/lib/roi_data_layer/roidb.py", line 116, in combined_roidb
roidbs = [get_roidb(s) for s in imdb_names.split('+')]
File "/home/code/zhoulinmao/faster-rcnn.pytorch/lib/roi_data_layer/roidb.py", line 109, in get_roidb
imdb = get_imdb(imdb_name)
File "/home/code/zhoulinmao/faster-rcnn.pytorch/lib/datasets/factory.py", line 66, in get_imdb
return __setsname
File "/home/code/zhoulinmao/faster-rcnn.pytorch/lib/datasets/factory.py", line 31, in
__sets[name] = (lambda split=split, year=year: coco(split, year))
File "/home/code/zhoulinmao/faster-rcnn.pytorch/lib/datasets/coco.py", line 38, in __init__
self._COCO = COCO(self._get_ann_file())
File "/home/code/zhoulinmao/faster-rcnn.pytorch/lib/pycocotools/coco.py", line 85, in __init__
dataset = json.load(open(annotation_file, 'r'))
IOError: [Errno 2] No such file or directory: '/home/code/zhoulinmao/faster-rcnn.pytorch/data/coco/annotations/instances_valminusminival2014.json'

The instances_valminusminival2014.json' is not exist in coco2014，how to do？

[ArchieGu]

@Markovcom

Hi, could you tell me how to prepare the COCO dataset?
I downloaded the COCO2014 train, test,val data form COCO official site and put them in to /data/coco .
I also put coco_proposals inside /data .
Besides create a symbolic link, what more should I do???
Thanks a lot.

[Dataphz]

@Markovcom As for me, i comment the code in factory.py as follow.

In factory.py:
for year in ['2014']:
for split in ['train', 'val']#, 'minival', 'valminusminival', 'trainval']:
name = 'coco_{}_{}'.format(year, split)
__sets[name] = (lambda split=split, year=year: coco(split, year))

for year in ['2014']:
for split in ['train', 'val']#, 'capval', 'valminuscapval', 'trainval']:
name = 'coco_{}_{}'.format(year, split)
__sets[name] = (lambda split=split, year=year: coco(split, year))

What's more, @ArchieGu about the data preparing, mkdir data/coco/images and data/coco/annotations, and put the images dataset in data/coco/images, put annotations in data/coco/annotations.