add missing preprocessing file

data/preprocess/preprocess_default.py

@@ -0,0 +1,49 @@
+#!/usr/bin/env python
+"""Preprocess a KGE dataset into a the format expected by libkge.
+
+Call as `preprocess_base.py --folder <name>`. The original dataset should be stored in
+subfolder `name` and have files "train.txt", "valid.txt", and "test.txt". Each file
+contains one field_map triple per line, separated by tabs.
+
+During preprocessing, each distinct entity name and each distinct relation name
+is assigned an index (dense). The index-to-object mapping is stored in files
+"entity_ids.del" and "relation_ids.del", resp. The triples (as indexes) are stored in
+files "train.del", "valid.del", and "test.del". Additionally, the splits
+"train_sample.del" (a random subset of train) and "valid_without_unseen.del" and
+"test_without_unseen.del" are stored. The "test/valid_without_unseen.del" files are
+subsets of "valid.del" and "test.del" resp. where all triples containing entities
+or relations not existing in "train.del" have been filtered out.
+
+Metadata information is stored in a file "dataset.yaml".
+
+"""
+
+import util
+
+if __name__ == "__main__":
+    args = util.default_parser().parse_args()
+    field_map = {
+        "S": args.subject_field,
+        "P": args.predicate_field,
+        "O": args.object_field,
+    }
+
+    print(f"Preprocessing {args.folder}...")
+
+    # register raw splits
+    train_raw = util.RawSplit(
+        file="train.txt",
+        field_map=field_map,
+        collect_entities=True,
+        collect_relations=True,
+    )
+    valid_raw = util.RawSplit(file="valid.txt", field_map=field_map,)
+    test_raw = util.RawSplit(file="test.txt", field_map=field_map,)
+
+    # create raw dataset with default splits
+    raw_dataset = util.create_raw_dataset(train_raw, valid_raw, test_raw, args)
+
+    # do the work
+    util.process_splits(raw_dataset)
+    util.update_string_files(raw_dataset, args)
+    util.write_dataset_yaml(raw_dataset.config, args.folder)