clean cartoonset

cartoonset.py

@@ -11,18 +11,32 @@ import datasets
 from datasets.tasks import ImageClassification
 
 
-_CITATION = """\
-@TECHREPORT{Krizhevsky09learningmultiple,
-    author = {Alex Krizhevsky},
-    title = {Learning multiple layers of features from tiny images},
-    institution = {},
-    year = {2009}
+_CITATION = r"""
+@article{DBLP:journals/corr/abs-1711-05139,
+  author    = {Amelie Royer and
+               Konstantinos Bousmalis and
+               Stephan Gouws and
+               Fred Bertsch and
+               Inbar Mosseri and
+               Forrester Cole and
+               Kevin Murphy},
+  title     = {{XGAN:} Unsupervised Image-to-Image Translation for many-to-many Mappings},
+  journal   = {CoRR},
+  volume    = {abs/1711.05139},
+  year      = {2017},
+  url       = {http://arxiv.org/abs/1711.05139},
+  eprinttype = {arXiv},
+  eprint    = {1711.05139},
+  timestamp = {Mon, 13 Aug 2018 16:47:38 +0200},
+  biburl    = {https://dblp.org/rec/journals/corr/abs-1711-05139.bib},
+  bibsource = {dblp computer science bibliography, https://dblp.org}
 }
 """
 
 _DESCRIPTION = """\
-The Cartoonset-10k dataset consists of 60000 32x32 colour images in 10 classes, with 6000 images
-per class. There are 50000 training images and 10000 test images.
+Cartoon Set is a collection of random, 2D cartoon avatar images. The cartoons vary in 10 artwork 
+categories, 4 color categories, and 4 proportion categories, with a total of ~1013 possible 
+combinations. We provide sets of 10k and 100k randomly chosen cartoons and labeled attributes. 
 """
 
 _DATA_URLS = {
@@ -45,7 +59,7 @@ class Cartoonset(datasets.GeneratorBasedBuilder):
         datasets.BuilderConfig(
             name="100k",
             version=datasets.Version("1.0.0", ""),
-            description="Loads the Cartoonset-10k Data Set",
+            description="Loads the Cartoonset-100k Data Set",
         ),
     ]
 
@@ -57,24 +71,16 @@ class Cartoonset(datasets.GeneratorBasedBuilder):
             features=datasets.Features(
                 {
                     "img": datasets.Image(),
-                    # "label": datasets.features.ClassLabel(names=_NAMES),
                 }
             ),
             supervised_keys=("img",),
             homepage="https://www.cs.toronto.edu/~kriz/cifar.html",
             citation=_CITATION,
-            # task_templates=ImageClassification(
-            #     image_column="img", label_column="label"
-            # ),
         )
 
     def _split_generators(self, dl_manager: datasets.DownloadManager):
         url = _DATA_URLS[self.config.name]
         print("URL:", url)
-        exit()
-        archive = dl_manager.download(url)
-
-        print(archive)
 
         return [
             datasets.SplitGenerator(
@@ -84,28 +90,17 @@ class Cartoonset(datasets.GeneratorBasedBuilder):
                     "split": "train",
                 },
             ),
-            # datasets.SplitGenerator(
-            #     name=datasets.Split.TEST, gen_kwargs={"files": dl_manager.iter_archive(archive), "split": "test"}
-            # ),
         ]
 
     def _generate_examples(self, files, split):
         """This function returns the examples in the raw (text) form."""
 
-        # if split == "train":
-        #     batches = ["data_batch_1", "data_batch_2", "data_batch_3", "data_batch_4", "data_batch_5"]
-
-        # if split == "test":
-        #     batches = ["test_batch"]
-        # batches = [f"Cartoonset-10k-batches-py/{filename}" for filename in batches]
-        print("FILES", files)
-
         path: str
         for path, file_obj in files:
 
             if path.endswith(".png"):
-                image = PIL.Image.open(path)
+                image = PIL.Image.open(file_obj)
 
                 yield path, {
-                    "img": np.asarray(image),
+                    "img": image,
                 }