MMDetction之配置文件：Faster RCNN resnet50 without FPN

ORIENT.

于 2024-05-24 22:01:26 发布

阅读量474

点赞数 5

CC 4.0 BY-SA版权

文章标签：深度学习人工智能神经网络 python pytorch object detection

本文链接：https://blog.youkuaiyun.com/ori3nt/article/details/139184072

官方所提供的配置文件都是带有FPN结构的，所以上传一个不带FPN结构的配置文件，实验结果还在测试。

# faster rcnn caffe dc5.py
# model settings
norm_cfg = dict(type='BN', requires_grad=False)
model = dict(
    type='FasterRCNN',
    data_preprocessor=dict(
        type='DetDataPreprocessor',
        mean=[103.530, 116.280, 123.675],
        std=[1.0, 1.0, 1.0],
        bgr_to_rgb=False,
        pad_size_divisor=32),
    backbone=dict(
        type='ResNet',
        depth=50,
        num_stages=4,
        strides=(1, 2, 2, 1),
        dilations=(1, 1, 1, 2),
        out_indices=(3, ),
        frozen_stages=1,
        norm_cfg=norm_cfg,
        norm_eval=True,
        style='caffe',
        init_cfg=dict(
            type='Pretrained',
            checkpoint='open-mmlab://detectron2/resnet50_caffe')),
    rpn_head=dict(
        type='RPNHead',
        in_channels=204