mask_rcnn_internimage_s_fpn_3x_coco.py

# --------------------------------------------------------
# InternImage
# Copyright (c) 2022 OpenGVLab
# Licensed under The MIT License [see LICENSE for details]
# --------------------------------------------------------
_base_ = [
    '../_base_/models/mask_rcnn_r50_fpn.py',
    '../_base_/datasets/coco_instance.py',
    '../_base_/schedules/schedule_3x.py',
    '../_base_/default_runtime.py'
]
pretrained = 'https://huggingface.co/OpenGVLab/InternImage/resolve/main/internimage_s_1k_224.pth'
model = dict(
    backbone=dict(
        _delete_=True,
        type='InternImage',
        core_op='DCNv3',
        channels=80,
        depths=[4, 4, 21, 4],
        groups=[5, 10, 20, 40],
        mlp_ratio=4.,
        drop_path_rate=0.3,
        norm_layer='LN',
        layer_scale=1.0,
        offset_scale=1.0,
        post_norm=True,
        with_cp=False,
        out_indices=(0, 1, 2, 3),
        init_cfg=dict(type='Pretrained', checkpoint=pretrained)),
    neck=dict(
        type='FPN',
        in_channels=[80, 160, 320, 640],
        out_channels=256,
        num_outs=5))
img_norm_cfg = dict(
    mean=[123.675, 116.28, 103.53], std=[58.395, 57.12, 57.375], to_rgb=True)
# augmentation strategy originates from DETR / Sparse RCNN
train_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='LoadAnnotations', with_bbox=True, with_mask=True),
    dict(type='RandomFlip', flip_ratio=0.5),
    dict(type='AutoAugment',
         policies=[
             [
                 dict(type='Resize',
                      img_scale=[(480, 1333), (512, 1333), (544, 1333), (576, 1333),
                                 (608, 1333), (640, 1333), (672, 1333), (704, 1333),
                                 (736, 1333), (768, 1333), (800, 1333)],
                      multiscale_mode='value',
                      keep_ratio=True)
             ],
             [
                 dict(type='Resize',
                      img_scale=[(400, 1333), (500, 1333), (600, 1333)],
                      multiscale_mode='value',
                      keep_ratio=True),
                 dict(type='RandomCrop',
                      crop_type='absolute_range',
                      crop_size=(384, 600),
                      allow_negative_crop=True),
                 dict(type='Resize',
                      img_scale=[(480, 1333), (512, 1333), (544, 1333),
                                 (576, 1333), (608, 1333), (640, 1333),
                                 (672, 1333), (704, 1333), (736, 1333),
                                 (768, 1333), (800, 1333)],
                      multiscale_mode='value',
                      override=True,
                      keep_ratio=True)
             ]
         ]),
    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']),
]
# By default, models are trained on 8 GPUs with 2 images per GPU
data = dict(
    samples_per_gpu=2,
    train=dict(pipeline=train_pipeline))
optimizer = dict(
    _delete_=True, type='AdamW', lr=0.0001, weight_decay=0.05,
    constructor='CustomLayerDecayOptimizerConstructor',
    paramwise_cfg=dict(num_layers=33, layer_decay_rate=1.0,
                       depths=[4, 4, 21, 4]))
optimizer_config = dict(grad_clip=None)
# fp16 = dict(loss_scale=dict(init_scale=512))
evaluation = dict(save_best='auto')
checkpoint_config = dict(
    interval=1,
    max_keep_ckpts=3,
    save_last=True,
)