Added splits

VQG.py

@@ -47,6 +47,12 @@ Visual questions for data science
 
 _URL = "https://huggingface.co/datasets/eduvedras/VQG/resolve/main/images.tar.gz"
 
+_METADATA_URLS = {
+                "train": "https://huggingface.co/datasets/eduvedras/VQG/resolve/main/metadata_train.csv", 
+                "validation": "https://huggingface.co/datasets/eduvedras/VQG/resolve/main/metadata_validation.csv",
+                "test": "https://huggingface.co/datasets/eduvedras/VQG/resolve/main/metadata_test.csv"
+            },
+
 class VQGTargz(datasets.GeneratorBasedBuilder):
     """SQUAD: The Stanford Question Answering Dataset. Version 1.1."""
 
@@ -75,11 +81,15 @@ class VQGTargz(datasets.GeneratorBasedBuilder):
     def _split_generators(self, dl_manager):
         path = dl_manager.download(_URL)
         image_iters = dl_manager.iter_archive(path)
-        metadata_path = dl_manager.download('https://huggingface.co/datasets/eduvedras/VQG/resolve/main/metadata.csv')
+        split_metadata_path = dl_manager.download(_METADATA_URLS)
 
         return [
             datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"images": image_iters,
-                                                                           "metadata_path": metadata_path}),
+                                                                           "metadata_path": split_metadata_path['train']}),
+            datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"images": image_iters,
+                                                                           "metadata_path": split_metadata_path['validation']}),
+            datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"images": image_iters,
+                                                                           "metadata_path": split_metadata_path['test']}),
         ]
 
     def _generate_examples(self, images, metadata_path):

metadata.csv

@@ -11703,101 +11703,6 @@ Breast_Cancer_histograms.png,Feature generation based on both variables smoothne
 Breast_Cancer_histograms.png,Feature generation based on both variables symmetry_se and perimeter_worst seems to be promising.,11701
 Breast_Cancer_histograms.png,Feature generation based on both variables radius_worst and perimeter_worst seems to be promising.,11702
 Breast_Cancer_histograms.png,Feature generation based on both variables texture_worst and perimeter_worst seems to be promising.,11703
-Breast_Cancer_mv.png,There is no reason to believe that discarding records showing missing values is safer than discarding the corresponding variables in this case.,11704
-Breast_Cancer_mv.png,Dropping all rows with missing values can lead to a dataset with less than 25% of the original data.,11705
-Breast_Cancer_mv.png,Dropping all rows with missing values can lead to a dataset with less than 30% of the original data.,11706
-Breast_Cancer_mv.png,Dropping all rows with missing values can lead to a dataset with less than 40% of the original data.,11707
-Breast_Cancer_mv.png,Dropping all records with missing values would be better than to drop the variables with missing values.,11708
-Breast_Cancer_mv.png,Discarding variables perimeter_mean and texture_mean would be better than discarding all the records with missing values for those variables.,11709
-Breast_Cancer_mv.png,Discarding variables texture_se and texture_mean would be better than discarding all the records with missing values for those variables.,11710
-Breast_Cancer_mv.png,Discarding variables perimeter_se and texture_mean would be better than discarding all the records with missing values for those variables.,11711
-Breast_Cancer_mv.png,Discarding variables area_se and texture_mean would be better than discarding all the records with missing values for those variables.,11712
-Breast_Cancer_mv.png,Discarding variables smoothness_se and texture_mean would be better than discarding all the records with missing values for those variables.,11713
-Breast_Cancer_mv.png,Discarding variables symmetry_se and texture_mean would be better than discarding all the records with missing values for those variables.,11714
-Breast_Cancer_mv.png,Discarding variables radius_worst and texture_mean would be better than discarding all the records with missing values for those variables.,11715
-Breast_Cancer_mv.png,Discarding variables texture_worst and texture_mean would be better than discarding all the records with missing values for those variables.,11716
-Breast_Cancer_mv.png,Discarding variables perimeter_worst and texture_mean would be better than discarding all the records with missing values for those variables.,11717
-Breast_Cancer_mv.png,Discarding variables texture_mean and perimeter_mean would be better than discarding all the records with missing values for those variables.,11718
-Breast_Cancer_mv.png,Discarding variables texture_se and perimeter_mean would be better than discarding all the records with missing values for those variables.,11719
-Breast_Cancer_mv.png,Discarding variables perimeter_se and perimeter_mean would be better than discarding all the records with missing values for those variables.,11720
-Breast_Cancer_mv.png,Discarding variables area_se and perimeter_mean would be better than discarding all the records with missing values for those variables.,11721
-Breast_Cancer_mv.png,Discarding variables smoothness_se and perimeter_mean would be better than discarding all the records with missing values for those variables.,11722
-Breast_Cancer_mv.png,Discarding variables symmetry_se and perimeter_mean would be better than discarding all the records with missing values for those variables.,11723
-Breast_Cancer_mv.png,Discarding variables radius_worst and perimeter_mean would be better than discarding all the records with missing values for those variables.,11724
-Breast_Cancer_mv.png,Discarding variables texture_worst and perimeter_mean would be better than discarding all the records with missing values for those variables.,11725
-Breast_Cancer_mv.png,Discarding variables perimeter_worst and perimeter_mean would be better than discarding all the records with missing values for those variables.,11726
-Breast_Cancer_mv.png,Discarding variables texture_mean and texture_se would be better than discarding all the records with missing values for those variables.,11727
-Breast_Cancer_mv.png,Discarding variables perimeter_mean and texture_se would be better than discarding all the records with missing values for those variables.,11728
-Breast_Cancer_mv.png,Discarding variables perimeter_se and texture_se would be better than discarding all the records with missing values for those variables.,11729
-Breast_Cancer_mv.png,Discarding variables area_se and texture_se would be better than discarding all the records with missing values for those variables.,11730
-Breast_Cancer_mv.png,Discarding variables smoothness_se and texture_se would be better than discarding all the records with missing values for those variables.,11731
-Breast_Cancer_mv.png,Discarding variables symmetry_se and texture_se would be better than discarding all the records with missing values for those variables.,11732
-Breast_Cancer_mv.png,Discarding variables radius_worst and texture_se would be better than discarding all the records with missing values for those variables.,11733
-Breast_Cancer_mv.png,Discarding variables texture_worst and texture_se would be better than discarding all the records with missing values for those variables.,11734
-Breast_Cancer_mv.png,Discarding variables perimeter_worst and texture_se would be better than discarding all the records with missing values for those variables.,11735
-Breast_Cancer_mv.png,Discarding variables texture_mean and perimeter_se would be better than discarding all the records with missing values for those variables.,11736
-Breast_Cancer_mv.png,Discarding variables perimeter_mean and perimeter_se would be better than discarding all the records with missing values for those variables.,11737
-Breast_Cancer_mv.png,Discarding variables texture_se and perimeter_se would be better than discarding all the records with missing values for those variables.,11738
-Breast_Cancer_mv.png,Discarding variables area_se and perimeter_se would be better than discarding all the records with missing values for those variables.,11739
-Breast_Cancer_mv.png,Discarding variables smoothness_se and perimeter_se would be better than discarding all the records with missing values for those variables.,11740
-Breast_Cancer_mv.png,Discarding variables symmetry_se and perimeter_se would be better than discarding all the records with missing values for those variables.,11741
-Breast_Cancer_mv.png,Discarding variables radius_worst and perimeter_se would be better than discarding all the records with missing values for those variables.,11742
-Breast_Cancer_mv.png,Discarding variables texture_worst and perimeter_se would be better than discarding all the records with missing values for those variables.,11743
-Breast_Cancer_mv.png,Discarding variables perimeter_worst and perimeter_se would be better than discarding all the records with missing values for those variables.,11744
-Breast_Cancer_mv.png,Discarding variables texture_mean and area_se would be better than discarding all the records with missing values for those variables.,11745
-Breast_Cancer_mv.png,Discarding variables perimeter_mean and area_se would be better than discarding all the records with missing values for those variables.,11746
-Breast_Cancer_mv.png,Discarding variables texture_se and area_se would be better than discarding all the records with missing values for those variables.,11747
-Breast_Cancer_mv.png,Discarding variables perimeter_se and area_se would be better than discarding all the records with missing values for those variables.,11748
-Breast_Cancer_mv.png,Discarding variables smoothness_se and area_se would be better than discarding all the records with missing values for those variables.,11749
-Breast_Cancer_mv.png,Discarding variables symmetry_se and area_se would be better than discarding all the records with missing values for those variables.,11750
-Breast_Cancer_mv.png,Discarding variables radius_worst and area_se would be better than discarding all the records with missing values for those variables.,11751
-Breast_Cancer_mv.png,Discarding variables texture_worst and area_se would be better than discarding all the records with missing values for those variables.,11752
-Breast_Cancer_mv.png,Discarding variables perimeter_worst and area_se would be better than discarding all the records with missing values for those variables.,11753
-Breast_Cancer_mv.png,Discarding variables texture_mean and smoothness_se would be better than discarding all the records with missing values for those variables.,11754
-Breast_Cancer_mv.png,Discarding variables perimeter_mean and smoothness_se would be better than discarding all the records with missing values for those variables.,11755
-Breast_Cancer_mv.png,Discarding variables texture_se and smoothness_se would be better than discarding all the records with missing values for those variables.,11756
-Breast_Cancer_mv.png,Discarding variables perimeter_se and smoothness_se would be better than discarding all the records with missing values for those variables.,11757
-Breast_Cancer_mv.png,Discarding variables area_se and smoothness_se would be better than discarding all the records with missing values for those variables.,11758
-Breast_Cancer_mv.png,Discarding variables symmetry_se and smoothness_se would be better than discarding all the records with missing values for those variables.,11759
-Breast_Cancer_mv.png,Discarding variables radius_worst and smoothness_se would be better than discarding all the records with missing values for those variables.,11760
-Breast_Cancer_mv.png,Discarding variables texture_worst and smoothness_se would be better than discarding all the records with missing values for those variables.,11761
-Breast_Cancer_mv.png,Discarding variables perimeter_worst and smoothness_se would be better than discarding all the records with missing values for those variables.,11762
-Breast_Cancer_mv.png,Discarding variables texture_mean and symmetry_se would be better than discarding all the records with missing values for those variables.,11763
-Breast_Cancer_mv.png,Discarding variables perimeter_mean and symmetry_se would be better than discarding all the records with missing values for those variables.,11764
-Breast_Cancer_mv.png,Discarding variables texture_se and symmetry_se would be better than discarding all the records with missing values for those variables.,11765
-Breast_Cancer_mv.png,Discarding variables perimeter_se and symmetry_se would be better than discarding all the records with missing values for those variables.,11766
-Breast_Cancer_mv.png,Discarding variables area_se and symmetry_se would be better than discarding all the records with missing values for those variables.,11767
-Breast_Cancer_mv.png,Discarding variables smoothness_se and symmetry_se would be better than discarding all the records with missing values for those variables.,11768
-Breast_Cancer_mv.png,Discarding variables radius_worst and symmetry_se would be better than discarding all the records with missing values for those variables.,11769
-Breast_Cancer_mv.png,Discarding variables texture_worst and symmetry_se would be better than discarding all the records with missing values for those variables.,11770
-Breast_Cancer_mv.png,Discarding variables perimeter_worst and symmetry_se would be better than discarding all the records with missing values for those variables.,11771
-Breast_Cancer_mv.png,Discarding variables texture_mean and radius_worst would be better than discarding all the records with missing values for those variables.,11772
-Breast_Cancer_mv.png,Discarding variables perimeter_mean and radius_worst would be better than discarding all the records with missing values for those variables.,11773
-Breast_Cancer_mv.png,Discarding variables texture_se and radius_worst would be better than discarding all the records with missing values for those variables.,11774
-Breast_Cancer_mv.png,Discarding variables perimeter_se and radius_worst would be better than discarding all the records with missing values for those variables.,11775
-Breast_Cancer_mv.png,Discarding variables area_se and radius_worst would be better than discarding all the records with missing values for those variables.,11776
-Breast_Cancer_mv.png,Discarding variables smoothness_se and radius_worst would be better than discarding all the records with missing values for those variables.,11777
-Breast_Cancer_mv.png,Discarding variables symmetry_se and radius_worst would be better than discarding all the records with missing values for those variables.,11778
-Breast_Cancer_mv.png,Discarding variables texture_worst and radius_worst would be better than discarding all the records with missing values for those variables.,11779
-Breast_Cancer_mv.png,Discarding variables perimeter_worst and radius_worst would be better than discarding all the records with missing values for those variables.,11780
-Breast_Cancer_mv.png,Discarding variables texture_mean and texture_worst would be better than discarding all the records with missing values for those variables.,11781
-Breast_Cancer_mv.png,Discarding variables perimeter_mean and texture_worst would be better than discarding all the records with missing values for those variables.,11782
-Breast_Cancer_mv.png,Discarding variables texture_se and texture_worst would be better than discarding all the records with missing values for those variables.,11783
-Breast_Cancer_mv.png,Discarding variables perimeter_se and texture_worst would be better than discarding all the records with missing values for those variables.,11784
-Breast_Cancer_mv.png,Discarding variables area_se and texture_worst would be better than discarding all the records with missing values for those variables.,11785
-Breast_Cancer_mv.png,Discarding variables smoothness_se and texture_worst would be better than discarding all the records with missing values for those variables.,11786
-Breast_Cancer_mv.png,Discarding variables symmetry_se and texture_worst would be better than discarding all the records with missing values for those variables.,11787
-Breast_Cancer_mv.png,Discarding variables radius_worst and texture_worst would be better than discarding all the records with missing values for those variables.,11788
-Breast_Cancer_mv.png,Discarding variables perimeter_worst and texture_worst would be better than discarding all the records with missing values for those variables.,11789
-Breast_Cancer_mv.png,Discarding variables texture_mean and perimeter_worst would be better than discarding all the records with missing values for those variables.,11790
-Breast_Cancer_mv.png,Discarding variables perimeter_mean and perimeter_worst would be better than discarding all the records with missing values for those variables.,11791
-Breast_Cancer_mv.png,Discarding variables texture_se and perimeter_worst would be better than discarding all the records with missing values for those variables.,11792
-Breast_Cancer_mv.png,Discarding variables perimeter_se and perimeter_worst would be better than discarding all the records with missing values for those variables.,11793
-Breast_Cancer_mv.png,Discarding variables area_se and perimeter_worst would be better than discarding all the records with missing values for those variables.,11794
-Breast_Cancer_mv.png,Discarding variables smoothness_se and perimeter_worst would be better than discarding all the records with missing values for those variables.,11795
-Breast_Cancer_mv.png,Discarding variables symmetry_se and perimeter_worst would be better than discarding all the records with missing values for those variables.,11796
-Breast_Cancer_mv.png,Discarding variables radius_worst and perimeter_worst would be better than discarding all the records with missing values for those variables.,11797
-Breast_Cancer_mv.png,Discarding variables texture_worst and perimeter_worst would be better than discarding all the records with missing values for those variables.,11798
 Breast_Cancer_histograms.png,The variable texture_mean can be coded as ordinal without losing information.,11799
 Breast_Cancer_histograms.png,The variable perimeter_mean can be coded as ordinal without losing information.,11800
 Breast_Cancer_histograms.png,The variable texture_se can be coded as ordinal without losing information.,11801