sticky-sandwich-47614
03/17/2021, 5:13 PMloud-glass-33663
03/18/2021, 5:17 AMbitter-boots-24291
03/18/2021, 9:29 AMkind-autumn-55765
03/19/2021, 4:18 AMhundreds-winter-88791
03/19/2021, 11:37 AMcold-nest-70029
03/19/2021, 12:02 PMcold-school-3088
03/19/2021, 1:28 PMdazzling-autumn-14365
03/19/2021, 7:33 PMeager-plastic-76453
03/20/2021, 9:45 AMcareful-pencil-58247
03/20/2021, 1:03 PMtall-advantage-5586
03/20/2021, 1:49 PMfaint-energy-11966
03/21/2021, 7:11 AMcalm-noon-73911
03/22/2021, 5:43 AMfaint-energy-11966
03/22/2021, 10:38 AMcalm-grass-85557
03/22/2021, 3:06 PMsticky-sandwich-47614
03/22/2021, 3:11 PMorange-action-12733
03/23/2021, 11:12 AMgorgeous-nail-79588
03/24/2021, 7:03 AMswift-london-82344
03/24/2021, 1:27 PMelegant-rose-70104
03/25/2021, 3:47 PMorange-action-12733
03/26/2021, 8:55 AMrefined-vr-28621
03/27/2021, 5:33 AMfaint-energy-11966
03/27/2021, 9:43 AMPyTorch?
I need to split the CIFAR10 dataset into training and validation set. The problem is that I wish to apply augmentations to training data. These are applied while loading the data. But if I split the data into validation set it also contains the augmentations which I obviously don't want
train_transform = transforms.Compose([transforms.RandomRotation(10),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
train_data = datasets.CIFAR10('data', train=True, download=True, transform=train_transform)
# Generate validation set - <https://stackoverflow.com/a/51768651|https://stackoverflow.com/a/51768651>
train_size = int(len(train_data) * 0.8) # 80% training data
valid_size = len(train_data) - train_size # 20% validation data
train_data, valid_data = random_split(train_data, [train_size, valid_size])
This snippet above applies augmentations to validation data (since it is just a part of the originally loaded data). Can I split the data before applying transforms?broad-magazine-64835
03/28/2021, 4:16 PMfaint-energy-11966
03/29/2021, 10:13 AMloud-glass-33663
03/31/2021, 1:05 PMcalm-grass-85557
03/31/2021, 1:32 PMswift-microphone-91058
04/01/2021, 12:08 PMloud-glass-33663
04/02/2021, 5:29 AM<mailto:groupname@provider.com|groupname@provider.com> which then delivers emails individually to all email addresses (with REPLY TO: my email).
Essentially, I want a mailchimp like functionality without having to go through the process of creating a campaign everytime. (I have a Google Apps account as well if there is a hack to get that going).
Any suggestions?
(Preferable to not have to bother the members to sign up / create an account / etc).acceptable-flag-71699
04/03/2021, 7:13 PM