Allow re-distribution of dev and test samples between splits

qanom.py

@@ -15,6 +15,8 @@
 """A Dataset loading script for the QANom dataset (klein et. al., COLING 2000)."""
 
 
+from dataclasses import dataclass
+from typing import Optional, Tuple
 import datasets
 from pathlib import Path
 import pandas as pd
@@ -71,6 +73,14 @@ _URLs = {
 
 SpanFeatureType = datasets.Sequence(datasets.Value("int32"), length=2)
 
+@dataclass
+class QANomBuilderConfig(datasets.BuilderConfig):
+    """ Allow the loader to re-distribute the original dev and test splits between train, dev and test. """
+    redistribute_dev: Tuple[float, float, float] = (0., 1., 0.)
+    redistribute_test: Tuple[float, float, float] = (0., 0., 1.)
+    
+    
+
 # TODO: Name of the dataset usually match the script name with CamelCase instead of snake_case
 class Qanom(datasets.GeneratorBasedBuilder):
     """QANom: Question-Answer driven SRL for Nominalizations corpus.
@@ -80,16 +90,18 @@ class Qanom(datasets.GeneratorBasedBuilder):
     are for canidate nominalization which are judged to be non-predicates ("is_verbal"==False) or predicates with no QAs.
     In these cases, the qa fields (question, answers, answer_ranges) would be empty lists. """
 
-    VERSION = datasets.Version("1.0.1")
+    VERSION = datasets.Version("1.0.2")
+    
+    BUILDER_CONFIG_CLASS = QANomBuilderConfig
 
     BUILDER_CONFIGS = [
-        datasets.BuilderConfig(
-            name="plain_text", version=VERSION, description="This provides the QANom dataset"
+        QANomBuilderConfig(
+            name="default", version=VERSION, description="This provides the QANom dataset"#, redistribute_dev=(0,1,0)
         ),
     ]
 
     DEFAULT_CONFIG_NAME = (
-        "plain_text"  # It's not mandatory to have a default configuration. Just use one if it make sense.
+        "default"  # It's not mandatory to have a default configuration. Just use one if it make sense.
     )
 
     def _info(self):
@@ -137,32 +149,54 @@ class Qanom(datasets.GeneratorBasedBuilder):
                 
     def _split_generators(self, dl_manager: datasets.utils.download_manager.DownloadManager):
         """Returns SplitGenerators."""            
-        
         # prepare wiktionary for verb inflections inside 'self.verb_inflections'
         self._prepare_wiktionary_verb_inflections(dl_manager)
         
-        corpus_base_path = Path(dl_manager.download_and_extract(_URLs["qanom_zip"]))        
+        self.corpus_base_path = Path(dl_manager.download_and_extract(_URLs["qanom_zip"]))        
+        
+        self.dataset_files = [
+            self.corpus_base_path / "annot.train.csv",
+            self.corpus_base_path / "annot.dev.csv",
+            self.corpus_base_path / "annot.test.csv"
+        ]
+        
+        # proportional segment (start,end) to take from every original split to returned SplitGenerator 
+        orig_dev_segments = ((0, self.config.redistribute_dev[0]),
+                             (self.config.redistribute_dev[0], sum(self.config.redistribute_dev[:2])),
+                             (sum(self.config.redistribute_dev[:2]), 1))
+        orig_tst_segments = ((0, self.config.redistribute_test[0]),
+                             (self.config.redistribute_test[0], sum(self.config.redistribute_test[:2])),
+                             (sum(self.config.redistribute_test[:2]), 1))
+        train_proportion = ((0,1),                  # from train
+                            orig_dev_segments[0],   # from dev
+                            orig_tst_segments[0])   # from test
+        dev_proportion = ((0,0),                    # from train
+                          orig_dev_segments[1],     # from dev
+                          orig_tst_segments[1])     # from test
+        test_proportion = ((0,0),                   # from train
+                           orig_dev_segments[2],    # from dev
+                           orig_tst_segments[2])    # from test
         
         return [
             datasets.SplitGenerator(
                 name=datasets.Split.TRAIN,
                 # These kwargs will be passed to _generate_examples
                 gen_kwargs={
-                    "filepath": corpus_base_path / "annot.train.csv",
+                    "split_proportion": train_proportion
                 },
             ),
             datasets.SplitGenerator(
                 name=datasets.Split.VALIDATION,
                 # These kwargs will be passed to _generate_examples
                 gen_kwargs={
-                    "filepath": corpus_base_path / "annot.dev.csv",
+                    "split_proportion": dev_proportion
                 },
             ),
             datasets.SplitGenerator(
                 name=datasets.Split.TEST,
                 # These kwargs will be passed to _generate_examples
                 gen_kwargs={
-                    "filepath": corpus_base_path / "annot.test.csv",
+                    "split_proportion": test_proportion
                 },
             ),
         ]
@@ -172,11 +206,17 @@ class Qanom(datasets.GeneratorBasedBuilder):
         start, end = s.split(":")
         return [int(start), int(end)]
     
-    def _generate_examples(self, filepath):
+    def _generate_examples(self, split_proportion=None):
 
         """ Yields examples from a 'annot.?.csv' file in QANom's format."""
 
-        df = pd.read_csv(filepath)
+        # construct concatenated DataFrame from different source splits
+        orig_splits_dfs = [pd.read_csv(filepath)
+                           for filepath in self.dataset_files]  # train, dev, test
+        segment_df_from_orig_splits = [df.iloc[int(len(df)*start) : int(len(df)*end)] 
+                                       for df, (start,end) in zip(orig_splits_dfs, split_proportion)] 
+            
+        df = pd.concat(segment_df_from_orig_splits, ignore_index=True)
         for counter, row in df.iterrows():
             # Each record (row) in csv is a QA or is stating a predicate/non-predicate with no QAs