Hi everyone,
Recently I tried to train my model on ImageNet
and I tried to use inception
and Alexnet
like preprocessing. I used Fast-ai imagenet training script. Pytorch
has support for inception
like preprocessing but for AlexNet
s Lighting
, we had to implement this one ourselves :
__imagenet_pca = {
'eigval': torch.Tensor([0.2175, 0.0188, 0.0045]),
'eigvec': torch.Tensor([
[-0.5675, 0.7192, 0.4009],
[-0.5808, -0.0045, -0.8140],
[-0.5836, -0.6948, 0.4203],
])
}
# Lighting data augmentation taken from here - https://github.com/eladhoffer/convNet.pytorch/blob/master/preprocess.py
class Lighting(object):
"""Lighting noise(AlexNet - style PCA - based noise)"""
def __init__(self, alphastd, eigval, eigvec):
self.alphastd = alphastd
self.eigval = eigval
self.eigvec = eigvec
def __call__(self, img):
if self.alphastd == 0:
return img
alpha = img.new().resize_(3).normal_(0, self.alphastd)
rgb = self.eigvec.type_as(img).clone()\
.mul(alpha.view(1, 3).expand(3, 3))\
.mul(self.eigval.view(1, 3).expand(3, 3))\
.sum(1).squeeze()
return img.add(rgb.view(3, 1, 1).expand_as(img))
which is finally used like this :
train_tfms = transforms.Compose([
transforms.RandomResizedCrop(size),
transforms.RandomHorizontalFlip(),
transforms.ColorJitter(.4,.4,.4),
transforms.ToTensor(),
Lighting(0.1, __imagenet_pca['eigval'], __imagenet_pca['eigvec']),
normalize,
])
train_dataset = datasets.ImageFolder(traindir, train_tfms)
train_sampler = (torch.utils.data.distributed.DistributedSampler(train_dataset)
if args.distributed else None)
train_loader = torch.utils.data.DataLoader(
train_dataset, batch_size=args.batch_size, shuffle=(train_sampler is None),
num_workers=args.workers, pin_memory=True, sampler=train_sampler)
However, the problem is , whenever I run the script I get the:
‘AttributeError: ‘Image’ object has no attribute ‘new’’
Which complains about this line :
alpha = img.new().resize_(3).normal_(0, self.alphastd)
What is wrong here ? what am I missing here? Can anyone please help me on this?