102 lines
3.7 KiB
Python
102 lines
3.7 KiB
Python
# Copyright (c) OpenMMLab. All rights reserved.
|
|
import copy
|
|
import warnings
|
|
|
|
from ..builder import CLASSIFIERS, build_backbone, build_head, build_neck
|
|
from ..utils.augment import Augments
|
|
from .base import BaseClassifier
|
|
|
|
|
|
@CLASSIFIERS.register_module()
|
|
class ImageClassifier(BaseClassifier):
|
|
|
|
def __init__(self,
|
|
backbone,
|
|
neck=None,
|
|
head=None,
|
|
pretrained=None,
|
|
train_cfg=None,
|
|
init_cfg=None):
|
|
super(ImageClassifier, self).__init__(init_cfg)
|
|
|
|
if pretrained is not None:
|
|
warnings.warn('DeprecationWarning: pretrained is a deprecated \
|
|
key, please consider using init_cfg')
|
|
self.init_cfg = dict(type='Pretrained', checkpoint=pretrained)
|
|
|
|
self.backbone = build_backbone(backbone)
|
|
|
|
if neck is not None:
|
|
self.neck = build_neck(neck)
|
|
|
|
if head is not None:
|
|
self.head = build_head(head)
|
|
|
|
self.augments = None
|
|
if train_cfg is not None:
|
|
augments_cfg = train_cfg.get('augments', None)
|
|
if augments_cfg is not None:
|
|
self.augments = Augments(augments_cfg)
|
|
else:
|
|
# Considering BC-breaking
|
|
mixup_cfg = train_cfg.get('mixup', None)
|
|
cutmix_cfg = train_cfg.get('cutmix', None)
|
|
assert mixup_cfg is None or cutmix_cfg is None, \
|
|
'If mixup and cutmix are set simultaneously,' \
|
|
'use augments instead.'
|
|
if mixup_cfg is not None:
|
|
warnings.warn('The mixup attribute will be deprecated. '
|
|
'Please use augments instead.')
|
|
cfg = copy.deepcopy(mixup_cfg)
|
|
cfg['type'] = 'BatchMixup'
|
|
# In the previous version, mixup_prob is always 1.0.
|
|
cfg['prob'] = 1.0
|
|
self.augments = Augments(cfg)
|
|
if cutmix_cfg is not None:
|
|
warnings.warn('The cutmix attribute will be deprecated. '
|
|
'Please use augments instead.')
|
|
cfg = copy.deepcopy(cutmix_cfg)
|
|
cutmix_prob = cfg.pop('cutmix_prob')
|
|
cfg['type'] = 'BatchCutMix'
|
|
cfg['prob'] = cutmix_prob
|
|
self.augments = Augments(cfg)
|
|
|
|
def extract_feat(self, img):
|
|
"""Directly extract features from the backbone + neck."""
|
|
x = self.backbone(img)
|
|
if self.with_neck:
|
|
x = self.neck(x)
|
|
return x
|
|
|
|
def forward_train(self, img, gt_label, **kwargs):
|
|
"""Forward computation during training.
|
|
|
|
Args:
|
|
img (Tensor): of shape (N, C, H, W) encoding input images.
|
|
Typically these should be mean centered and std scaled.
|
|
gt_label (Tensor): It should be of shape (N, 1) encoding the
|
|
ground-truth label of input images for single label task. It
|
|
shoulf be of shape (N, C) encoding the ground-truth label
|
|
of input images for multi-labels task.
|
|
Returns:
|
|
dict[str, Tensor]: a dictionary of loss components
|
|
"""
|
|
if self.augments is not None:
|
|
img, gt_label = self.augments(img, gt_label)
|
|
|
|
x = self.extract_feat(img)
|
|
|
|
losses = dict()
|
|
loss = self.head.forward_train(x, gt_label)
|
|
losses.update(loss)
|
|
|
|
return losses
|
|
|
|
def simple_test(self, img, img_metas):
|
|
"""Test without augmentation."""
|
|
x = self.extract_feat(img)
|
|
x_dims = len(x.shape)
|
|
if x_dims == 1:
|
|
x.unsqueeze_(0)
|
|
return self.head.simple_test(x)
|