1. 训练faster-rcnn:
2. 训练cascade mask-rcnn:
修改 cascade_mask_rcnn_r50_fpn.py文件中所有的num_classes为你要训练的数据集的类别(不包括背景)
然后在configs/cascade_rcnn/cascade_rcnn_r50_fpn_1x_coco.py中添加数据集的设置
_base_ = [
'../_base_/models/cascade_mask_rcnn_r50_fpn.py',
'../_base_/datasets/coco_instance.py',
'../_base_/schedules/schedule_1x.py', '../_base_/default_runtime.py'
]
# 修改数据集相关设置
dataset_type = 'COCODataset'
classes = ('crack', 'hole',)
data = dict(
train=dict(
img_prefix='E:/jt/mmdetection/data/coco/defect_dataset/train/',
classes=classes,
ann_file='E:/jt/mmdetection/data/coco/defect_dataset/train/annotations_coco.json'),
val=dict