Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

How to use my own dataset #18

Open
hanshuai119 opened this issue Nov 20, 2018 · 8 comments
Open

How to use my own dataset #18

hanshuai119 opened this issue Nov 20, 2018 · 8 comments

Comments

@hanshuai119
Copy link

No description provided.

@ArghyaPal
Copy link

Hi,

open the dataloader.py file
and, add

elif dataset == 'AWA':
data_loader = DataLoader(
datasets.ImageFolder('data/AWA', transform=transform),
batch_size=batch_size, shuffle=True)

then, open main.py
and modify this line:
parser.add_argument('--dataset', type=str, default='mnist', choices=[ 'mnist', 'fashion-mnist', 'cifar10', 'cifar100', 'svhn', 'stl10', 'lsun-bed'],

to, parser.add_argument('--dataset', type=str, default='mnist', choices=['AWA', 'mnist', 'fashion-mnist', 'cifar10', 'cifar100', 'svhn', 'stl10', 'lsun-bed'],

@HamzahNizami
Copy link

Do the images have to be in a specific format or size.
Thanks in advance

@HamzahNizami
Copy link

Hi,

open the dataloader.py file
and, add

elif dataset == 'AWA':
data_loader = DataLoader(
datasets.ImageFolder('data/AWA', transform=transform),
batch_size=batch_size, shuffle=True)

then, open main.py
and modify this line:
parser.add_argument('--dataset', type=str, default='mnist', choices=[ 'mnist', 'fashion-mnist', 'cifar10', 'cifar100', 'svhn', 'stl10', 'lsun-bed'],

to, parser.add_argument('--dataset', type=str, default='mnist', choices=['AWA', 'mnist', 'fashion-mnist', 'cifar10', 'cifar100', 'svhn', 'stl10', 'lsun-bed'],

Do the images have to be in a specific format or size.
Thanks in advance

@ArghyaPal
Copy link

No, I don't think so. Please go ahead and use your dataset.. please reply back, if you encounter any problem related to image size

@HamzahNizami
Copy link

Okay I will try soon.

I have a dataset of my own JPG images. These images are of strawberries. The image I want to generate is of strawberries.

My images are in colour, will this affect training on this GitHub. Do I need to change to gray-scale.

Finally @ArghyaPal have you successfully trained this model on your dataset? Any help would be greatly appreciated.

Thanks in advance!

@HamzahNizami
Copy link

No, I don't think so. Please go ahead and use your dataset.. please reply back, if you encounter any problem related to image size

@ArghyaPal

Hi, I'm currently getting the following error.

image

Even though my images are jpg as shown in the image below. Any help would be appreicated.

image

@HamzahNizami
Copy link

I managed to fix this issue by placing them in another folder called 1, so my directory is now, data/strawberriesBlack/1/B(1).jpg

However, my results are completely blank after 50 epochs on GAN

image

My dataset is currently coloured, do you think I should change to grayscale?
How big should said dataset actually be?

Thanks in advance

@evdelph
Copy link

evdelph commented Feb 17, 2020

@ArghyaPal I am having issues with either size or color scaling with these models, . I am trying to run color images of ships through GAN but I get shape errors (see issue 25). Is there a fix to this?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants