mmdetectionv1.0.0-选择几张卡训练一个模型，单机多卡开多个训练任务(还有问题)---修改config-batch图像数目，学习率下降等--训练期间测试开启-训练灰度
如果你想训练灰度图，在这个版本，你应该：
mmdetection/mmdet/datasets/pipelines/loading.py
@PIPELINES.register_module
class LoadImageFromFile(object):
def __init__(self, to_float32=False, color_type='color'):
self.to_float32 = to_float32
self.color_type = color_type
def __call__(self, results):
if results['img_prefix'] is not None:
filename = osp.join(results['img_prefix'],
results['img_info']['filename'])
else:
filename = results['img_info']['filename']
img = mmcv.imread(filename, self.color_type)
if self.to_float32:
img = img.astype(np.float32)
results['filename'] = filename
results['img'] = img
results['img_shape'] = img.shape
results['ori_shape'] = img.shape
return results
def __repr__(self):
return '{} (to_float32={}, color_type={})'.format(
self.__class__.__name__, self.to_float32, self.color_type)
img = mmcv.imread(filename, self.color_type) 后面接一段代码：
import cv2
img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
img = cv2.cvtColor(img, cv2.COLOR_GRAY2BGR)
训练期间测试开启
--validate
#!/usr/bin/env bash
PYTHON=/home/apple/anaconda3/envs/py36_mmdetection/bin/python
CONFIG=$1
GPUS=$2
GPUNAME=$3
PORT=${PORT:-29500}
$PYTHON -m torch.distributed.launch --nproc_per_node=$GPUS --master_port=$PORT \
$(dirname "$0")/train.py $CONFIG --gpus $GPUNAME --validate --launcher pytorch ${@:4}
修改多少次进行一次测试：
configs/mask_rcnn_r50_fpn_1x.py
# yapf:enable
evaluation = dict(interval=50)
修改config-batch图像数目，学习率下降等
修改：
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
dict(type='Resize', img_scale=(1333, 800), keep_ratio=True),
dict(type='RandomFlip', flip_ratio=0.5),
dict(type='Normalize', **img_norm_cfg),
dict(type='Pad', size_divisor=32),
dict(type='DefaultFormatBundle'),
dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels', 'gt_masks']),
]
test_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='MultiScaleFlipAug',
img_scale=(1333, 800),
flip=False,
transforms=[
dict(type='Resize', keep_ratio=True),
dict(type='RandomFlip'),
dict(type='Normalize', **img_norm_cfg),
dict(type='Pad', size_divisor=32),
dict(type='ImageToTensor', keys=['img']),
dict(type='Collect', keys=['img']),
])
]
..............................
data = dict(
imgs_per_gpu=4,
workers_per_gpu=2,
train=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_train2017.json',
img_prefix=data_root + 'train2017/',
pipeline=train_pipeline),
val=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
pipeline=test_pipeline),
test=dict(
type=dataset_type,
ann_file=data_root + 'annotations/instances_val2017.json',
img_prefix=data_root + 'val2017/',
pipeline=test_pipeline))
..............................
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=1.0 / 3,
step=[40, 80, 120])
52% 71C P2 185W / 250W | 10227MiB / 11019MiB | 72% Default
显存刚好利用上，再大一点就不行了
https://github.com/open-mmlab/mmdetection/tree/v1.0.0
选择几张卡训练一个模型，单机多卡开多个训练任务
使用命令：
./tools/dist_train.sh configs/mask_rcnn_r50_fpn_1x.py 1 2
需要改代码：（--resume_from checkpoint/mask_rcnn_r50_fpn_1x_20181010-069fa190.pth 是我自己添加的，你跟据你的看情况，我的项目这样训练收敛快）
mmdetection/tools/dist_train.sh
#!/usr/bin/env bash
PYTHON=/root/train/results/new/anaconda3_py3.7/bin/python
CONFIG=$1
GPUS=$2
GPUNAME=$3
PORT=${PORT:-29502}
nohup $PYTHON -m torch.distributed.launch --nproc_per_node=$GPUS --master_port=$PORT \
$(dirname "$0")/train.py $CONFIG --resume_from checkpoint/mask_rcnn_r50_fpn_1x_20181010-069fa190.pth --gpus $GPUNAME --launcher pytorch ${@:4} > out_marker.log 2>&1 &
mmdetection/tools/train.py
def parse_args():
parser = argparse.ArgumentParser(description='Train a detector')
parser.add_argument('config', help='train config file path')
parser.add_argument('--work_dir', help='the dir to save logs and models')
parser.add_argument(
'--resume_from', help='the checkpoint file to resume from')
parser.add_argument(
'--validate',
action='store_true',
help='whether to evaluate the checkpoint during training')
parser.add_argument(
'--gpus',
#type=int, 注释掉
default=1,
help='number of gpus to use '
'(only applicable to non-distributed training)')
parser.add_argument('--seed', type=int, default=None, help='random seed')
parser.add_argument(
'--deterministic',
action='store_true',
help='whether to set deterministic options for CUDNN backend.')
parser.add_argument(
'--launcher',
choices=['none', 'pytorch', 'slurm', 'mpi'],
default='none',
help='job launcher')
parser.add_argument('--local_rank', type=int, default=0)
parser.add_argument(
'--autoscale-lr',
action='store_true',
help='automatically scale lr with the number of gpus')
args = parser.parse_args()
if 'LOCAL_RANK' not in os.environ:
os.environ['LOCAL_RANK'] = str(args.local_rank)
return args
def main():
args = parse_args()
cfg = Config.fromfile(args.config)
# set cudnn_benchmark
if cfg.get('cudnn_benchmark', False):
torch.backends.cudnn.benchmark = True
# update configs according to CLI args
if args.work_dir is not None:
cfg.work_dir = args.work_dir
if args.resume_from is not None:
cfg.resume_from = args.resume_from
#cfg.gpus = args.gpus 注释然后添加如下：
cfg.gpus = str(args.gpus).split(',')
if(len(cfg.gpus))==1:
cfg.gpus = int(cfg.gpus[0])
os.environ['CUDA_VISIBLE_DEVICES'] = str(args.gpus)
if args.autoscale_lr:
# apply the linear scaling rule (https://arxiv.org/abs/1706.02677)
cfg.optimizer['lr'] = cfg.optimizer['lr'] * cfg.gpus / 8
mmdetection/mmdet/api/train.py
def _non_dist_train(model,
dataset,
cfg,
validate=False,
logger=None,
timestamp=None):
if validate:
raise NotImplementedError('Built-in validation is not implemented '
'yet in not-distributed training. Use '
'distributed training or test.py and '
'*eval.py scripts instead.')
# prepare data loaders
dataset = dataset if isinstance(dataset, (list, tuple)) else [dataset]
data_loaders = [
build_dataloader(
ds,
cfg.data.imgs_per_gpu,
cfg.data.workers_per_gpu,
#cfg.gpus,
len(cfg.gpus),
dist=False) for ds in dataset
]
# put model on gpus
#model = MMDataParallel(model, device_ids=range(cfg.gpus)).cuda()
model = MMDataParallel(model, device_ids=cfg.gpus).cuda()
作者：知识在于分享
