CUB_200_2011 FGCV baseline

pretrain model

resnest50 -> 75.58%

数据增强方法:

transforms.Resize((self.BASE_RESIZE_SIZE, self.BASE_RESIZE_SIZE), Image.BILINEAR),
                transforms.RandomCrop(self.INPUT_SIZE),
                transforms.RandomHorizontalFlip(),
                transforms.RandomRotation(degrees=15),
                transforms.ColorJitter(brightness=self.BRIGHTNESS, contrast=self.CONTRAST, hue=self.HUE, saturation=self.SATURATION),
                transforms.ToTensor(),
                self.imagenet_normalization_paramters

学习率:

0.1初始化，不进行warmup 90个epoch，0.3, 0.6, 0.9 * epoch 按ratio=0.1进行线性衰减

performance

model	accuracy
resne50(baseline)256->224	76.47%
resnet50(baseline)512->448	84.55%
efnet-b0 256->224	77.93%
efnet-b1 272->240	79.63%
efnet-b2 292->260	81.22%
efnet-b3 352->300	83.12%
efnet-b4 416->380	84.45%
efnet-b5 512->456	85.64%
efnet-b6 600->528	87.05%
efnet-b7 640->600	88.11%

training

bash train.sh 其中，horovod可以指定几个node，对应的就是几块GPU

test

CUDA_VISIBLE_DEVICES=0 python test.py

pretrain download link

cub_resnet50_08456

accvbaseline

model	test acc
efnetb2 260	41.975
resnet50 224	36.11
efnetb2+resnet50_224 ensemble1	43.105
efnetb3 300	42.395
resnet50 448	41.63
efnetb3+resnet50_448 ensemble2	47.05
ensemble1+ensemble2	47.395

• 20201016 update

model	test acc	trick
efnetb5 456	51.04	cosine_lr+cutmix+40epoch+no_warmup+32bs+0.1lr
r50 448	47.64	cosine_lr+cutmix+40epoch+no_warmup+48bs+0.1lr

• 20201018 update

model	test acc	trick
regnet12gf	53.578	cosine_lr+cutmix+40epoch+no_warmup+48bs+0.1lr
regnet12gf + efnetb5	55.758	same