Where are datasets shown in readme?

[Zardinality]

There are sunflower2daisy and leopard2tiger results shown in readme, but I don't find any clue about where to download them in code. In https://github.com/luoxier/CycleGAN_Tensorlayer/blob/master/main.py#L32 an optional value for dataset_dir is sunflower2daisy, where can I get it? The author of original paper doesn't seem to provide it.

[zsdonghao]

check tl.file.load_cyclegan_dataset

[Zardinality]

@zsdonghao For me tl.file.load_cyclegan_datasetseems to function like downloading dataset from https://people.eecs.berkeley.edu/~taesung_park/CycleGAN/datasets/, the url which original aurhor provided. But the additional dataset like the sunflower2daisy is not included.

[zsdonghao]

i can't open the link you provide, but it can download all official dataset the cyclegan authors provide in his website.

[Zardinality]

@zsdonghao Sorry I mistoke the url with the label in markdown grammar. This is the right url: official dataset. Of course you could download those official dataset, but some results in readme are not official provided, like sunflower2daisy. My question is where can I download them and replicate relating results.

[zsdonghao]

em, we may need the author to answer it~

[luoxier]

I upload the dataset of sunflower2daisy now. This dataset is collected from Flickr by Bulkr and if you want to try more examples you can download them from Flickr by different tags.

[Zardinality]

@luoxier thanks for your efforts for managing an dataset! But I don't think it is a good idea to include it in repository, since it doesn't belongs to source codes that should to be kept in git. Moreover, users who are uninterested in this dataset are forced to download the whole dataset if they want to clone this repo. I have uploaded this dataset to google drive, I think it should be fine to included it in readme.