第五讲 MMPretrain代码课

xcy04556

已于 2023-06-07 00:38:10 修改

阅读量132

点赞数

分类专栏： AI实战营文章标签： python 数学建模开发语言

于 2023-06-06 22:42:45 首次发布

本文链接：https://blog.csdn.net/weixin_39109575/article/details/131076913

版权

AI实战营专栏收录该内容

12 篇文章 0 订阅

订阅专栏

基于Resnet18微调分类网络

配置文件：‘configs/resnet/resnet18_8xb32_in1k.py’

_base_ = [
    '../_base_/models/resnet18.py', '../_base_/datasets/imagenet_bs32.py',
    '../_base_/schedules/imagenet_bs256.py', '../_base_/default_runtime.py'
]

网络配置：‘open-mmlab/mmpretrain/configs/_base_/models/resnet18.py’

# model settings
model = dict(
    type='ImageClassifier',
    backbone=dict( #主干网络
        type='ResNet',
        depth=18,
        num_stages=4,
        out_indices=(3, ),
        style='pytorch'),
    neck=dict(type='GlobalAveragePooling'),
    head=dict(     #分类头
        type='LinearClsHead',
        num_classes=1000,
        in_channels=512,
        loss=dict(type='CrossEntropyLoss', loss_weight=1.0),
        topk=(1, 5),
    ))

数据集配置：‘configs/_base_/datasets/imagenet_bs32.py’

# dataset settings
dataset_type = 'ImageNet'
data_preprocessor = dict(
    num_classes=1000,
    # RGB format normalization parameters
    mean=[123.675, 116.28, 103.53],
    std=[58.395, 57.12, 57.375],
    # convert image from BGR to RGB
    to_rgb=True,
)

train_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='RandomResizedCrop', scale=224),
    dict(type='RandomFlip', prob=0.5, direction='horizontal'),
    dict(type='PackInputs'),
]

test_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='ResizeEdge', scale=256, edge='short'),
    dict(type='CenterCrop', crop_size=224),
    dict(type='PackInputs'),
]

train_dataloader = dict(
    batch_size=32,
    num_workers=5,
    dataset=dict(
        type=dataset_type,
        data_root='data/imagenet',
        ann_file='meta/train.txt',
        data_prefix='train',
        pipeline=train_pipeline),
    sampler=dict(type='DefaultSampler', shuffle=True),
)

val_dataloader = dict(
    batch_size=32,
    num_workers=5,
    dataset=dict(
        type=dataset_type,
        data_root='data/imagenet',
        ann_file='meta/val.txt',
        data_prefix='val',
        pipeline=test_pipeline),
    sampler=dict(type='DefaultSampler', shuffle=False),
)
val_evaluator = dict(type='Accuracy', topk=(1, 5))

# If you want standard test, please manually configure the test dataset
test_dataloader = val_dataloader
test_evaluator = val_evaluator

规划配置：‘configs/base/schedules/imagenet_bs256.py’

# optimizer
optim_wrapper = dict(
    optimizer=dict(type='SGD', lr=0.1, momentum=0.9, weight_decay=0.0001))

# learning policy
param_scheduler = dict(
    type='MultiStepLR', by_epoch=True, milestones=[30, 60, 90], gamma=0.1)

# train, val, test setting
train_cfg = dict(by_epoch=True, max_epochs=100, val_interval=1)
val_cfg = dict()
test_cfg = dict()

# NOTE: `auto_scale_lr` is for automatically scaling LR,
# based on the actual training batch size.
auto_scale_lr = dict(base_batch_size=256)

运行配置：‘configs/base/default_runtime.py’（一般默认）

# defaults to use registries in mmpretrain
default_scope = 'mmpretrain'

# configure default hooks
default_hooks = dict(
    # record the time of every iteration.
    timer=dict(type='IterTimerHook'),

    # print log every 100 iterations.
    logger=dict(type='LoggerHook', interval=100),

    # enable the parameter scheduler.
    param_scheduler=dict(type='ParamSchedulerHook'),

    # save checkpoint per epoch.
    checkpoint=dict(type='CheckpointHook', interval=1， max_keep_ckpts=5, save_best='auto'),

    # set sampler seed in distributed evrionment.
    sampler_seed=dict(type='DistSamplerSeedHook'),

    # validation results visualization, set True to enable it.
    visualization=dict(type='VisualizationHook', enable=False),
)

# configure environment
env_cfg = dict(
    # whether to enable cudnn benchmark
    cudnn_benchmark=False,

    # set multi process parameters
    mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0),

    # set distributed parameters
    dist_cfg=dict(backend='nccl'),
)

# set visualizer
vis_backends = [dict(type='LocalVisBackend')]
visualizer = dict(type='UniversalVisualizer', vis_backends=vis_backends)

# set log level
log_level = 'INFO'

# load from which checkpoint
load_from = None

# whether to resume training from the loaded checkpoint
resume = False

# Defaults to use random seed and disable `deterministic`
randomness = dict(seed=None, deterministic=False)

新建分类猫狗的配置文件：‘resnet18_finetune.py’

# model settings
model = dict(
    type='ImageClassifier',
    backbone=dict( #主干网络
        type='ResNet',
        depth=18,
        num_stages=4,
        out_indices=(3, ),
        style='pytorch'),
    neck=dict(type='GlobalAveragePooling'),
    head=dict(     #分类头
        type='LinearClsHead',
        num_classes=2, #修改分类数
        in_channels=512,
        loss=dict(type='CrossEntropyLoss', loss_weight=1.0),
        topk=(1, 5),
    ),
    #设置预训练权重
    init_cfg=dict(type='Pretrained', checkpoint='https://download.openmmlab.com/mmclassification/0/resnet/resnet18_8xb32-in1k-20210831-fbbb1da6.pth')

# dataset settings
dataset_type = 'CustomDataset'
data_preprocessor = dict(
    num_classes=1000,
    # RGB format normalization parameters
    mean=[123.675, 116.28, 103.53],
    std=[58.395, 57.12, 57.375],
    # convert image from BGR to RGB
    to_rgb=True,
)

train_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='RandomResizedCrop', scale=224),
    dict(type='RandomFlip', prob=0.5, direction='horizontal'),
    dict(type='PackInputs'),
]

test_pipeline = [
    dict(type='LoadImageFromFile'),
    dict(type='ResizeEdge', scale=256, edge='short'),
    dict(type='CenterCrop', crop_size=224),
    dict(type='PackInputs'),
]

train_dataloader = dict(
    batch_size=32,
    num_workers=5,
    dataset=dict(
        type=dataset_type,
        data_root='../../data/cat_dogs_dataset/training_set',
        pipeline=train_pipeline),
    sampler=dict(type='DefaultSampler', shuffle=True),
)

val_dataloader = dict(
    batch_size=32,
    num_workers=5,
    dataset=dict(
        type=dataset_type,
        data_root='../../data/cat_dogs_dataset/val_set',
        pipeline=test_pipeline),
    sampler=dict(type='DefaultSampler', shuffle=False),
)
val_evaluator = dict(type='Accuracy', topk=1)

# If you want standard test, please manually configure the test dataset
test_dataloader = dict(
    batch_size=32,
    num_workers=5,
    dataset=dict(
        type=dataset_type,
        data_root='../../data/cat_dogs_dataset/test_set',
        pipeline=test_pipeline),
    sampler=dict(type='DefaultSampler', shuffle=False),
)
test_evaluator = val_evaluator

# optimizer
optim_wrapper = dict(
    optimizer=dict(type='SGD', lr=0.1, momentum=0.9, weight_decay=0.0001))

# learning policy
param_scheduler = dict(
    type='MultiStepLR', by_epoch=True, milestones=[30, 60, 90], gamma=0.1)

# train, val, test setting
train_cfg = dict(by_epoch=True, max_epochs=100, val_interval=1)
val_cfg = dict()
test_cfg = dict()

# NOTE: `auto_scale_lr` is for automatically scaling LR,
# based on the actual training batch size.
auto_scale_lr = dict(base_batch_size=256)

# defaults to use registries in mmpretrain
default_scope = 'mmpretrain'

# configure default hooks
default_hooks = dict(
    # record the time of every iteration.
    timer=dict(type='IterTimerHook'),

    # print log every 100 iterations.
    logger=dict(type='LoggerHook', interval=100),

    # enable the parameter scheduler.
    param_scheduler=dict(type='ParamSchedulerHook'),

    # save checkpoint per epoch.
    checkpoint=dict(type='CheckpointHook', interval=1， max_keep_ckpts=5, save_best='auto'),

    # set sampler seed in distributed evrionment.
    sampler_seed=dict(type='DistSamplerSeedHook'),

    # validation results visualization, set True to enable it.
    visualization=dict(type='VisualizationHook', enable=False),
)

# configure environment
env_cfg = dict(
    # whether to enable cudnn benchmark
    cudnn_benchmark=False,

    # set multi process parameters
    mp_cfg=dict(mp_start_method='fork', opencv_num_threads=0),

    # set distributed parameters
    dist_cfg=dict(backend='nccl'),
)

# set visualizer
vis_backends = [dict(type='LocalVisBackend')]
visualizer = dict(type='UniversalVisualizer', vis_backends=vis_backends)

# set log level
log_level = 'INFO'

# load from which checkpoint
load_from = None

# whether to resume training from the loaded checkpoint
resume = False

# Defaults to use random seed and disable `deterministic`
randomness = dict(seed=None, deterministic=False)

训练

mim train mmpretrain resnet18_finetune.py --work-dir=./exp

测试

mim test mmpretrain resnet18_finetune.py --checkpoint exp/epoch_5.pth --out result.pkl

分析

mim run mmpretrain analyze_results resnet18_finetune.py result.pkl --out-dir analyze
mim run mmpretrain confusion_matrix resnet18_finetune.py result.pk] --show --include-values

推理

from mmpretrain import ImageClassificationInferencer
inferencer = ImageClassificationInferencer(',/resnet18-finetune.py', pretrained='exo/epoch 5.pth')
inferencer("/home/PJLAB/mazerun/Pictures/cat.jpeg"，show=True)