Added creation scripts

dataset_creation_scripts/annotate.py

@@ -0,0 +1,138 @@
+import json
+import pickle
+from pathlib import Path
+
+from spacy.tokens import Span
+
+import dacy
+from dacy.datasets import dane
+
+
+def load_examples():
+    save_path = Path("examples.pkl")
+    if save_path.exists():
+        with open(save_path, "rb") as f:
+            examples = pickle.load(f)
+
+        return examples
+
+    train, dev, test = dane()
+
+    nlp = dacy.load("da_dacy_large_ner_fine_grained-0.1.0")
+
+    examples = list(test(nlp)) + list(train(nlp)) + list(dev(nlp))
+
+    docs = nlp.pipe([ex.x.text for ex in examples])
+
+    for e in examples:
+        e.predicted = next(docs)
+
+    with open("examples.pkl", "wb") as f:
+        pickle.dump(examples, f)
+
+    return examples
+
+
+def normalize_examples(examples):
+    label_mapping = {
+        "PER": "PERSON",
+        "LOC": "LOCATION",
+        "ORG": "ORGANIZATION",
+        "MISC": "MISC",
+    }
+
+    for e in examples:
+        old_ents = e.y.ents
+        new_ents = []
+        for ent in old_ents:
+            new_label = label_mapping[ent.label_]
+            new_ent = Span(e.y, start=ent.start, end=ent.end, label=new_label)
+            new_ents.append(new_ent)
+
+        e.y.ents = new_ents
+
+    return examples
+
+
+def example_to_review_format(example) -> dict:
+    ref = example.y
+
+    text = ref.text
+    tokens = [
+        {"text": t.text, "start": t.idx, "end": t.idx + len(t), "id": i}
+        for i, t in enumerate(ref)
+    ]
+    answer = "accept"
+
+    versions = []
+
+    v_ref_spans = [
+        {
+            "start": s.start_char,
+            "end": s.end_char,
+            "label": s.label_,
+            "token_start": s.start,
+            "token_end": s.end - 1,
+        }
+        for s in ref.ents
+    ]
+    v_ref = {
+        "text": text,
+        "tokens": tokens,
+        "spans": v_ref_spans,
+        "answer": answer,
+        "sessions": ["reference"],
+        "default": True,
+    }
+    versions.append(v_ref)
+
+    v_pred_spans = [
+        {
+            "start": s.start_char,
+            "end": s.end_char,
+            "label": s.label_,
+            "token_start": s.start,
+            "token_end": s.end - 1,
+        }
+        for s in example.predicted.ents
+    ]
+    v_pred = {
+        "text": text,
+        "tokens": tokens,
+        "spans": v_pred_spans,
+        "answer": answer,
+        "sessions": ["da_dacy_large_ner_fine_grained-0.1.0"],
+        "default": True,
+    }
+    versions.append(v_pred)
+
+    return {
+        "text": text,
+        "tokens": tokens,
+        "answer": answer,
+        "view_id": "ner_manual",
+        "versions": versions,
+    }
+
+
+if __name__ == "__main__":
+    examples = load_examples()
+    
+    ",".join(set([ent.label_ for e in examples for ent in e.x.ents]))
+
+    jsonl_data = [example_to_review_format(e) for e in normalize_examples(examples)]
+
+    with open("examples.jsonl", "w") as f:
+        for json_dict in jsonl_data:
+            line = json.dumps(json_dict)
+            f.write(f"{line}\n")
+
+    with open("reference.jsonl", "w") as f:
+        for json_dict in jsonl_data:
+            line = json.dumps(json_dict["versions"][0])
+            f.write(f"{line}\n")
+
+    with open("predictions.jsonl", "w") as f:
+        for json_dict in jsonl_data:
+            line = json.dumps(json_dict["versions"][1])
+            f.write(f"{line}\n")

dataset_creation_scripts/split.py

@@ -0,0 +1,118 @@
+import spacy
+from numpy import char
+from spacy.tokens import Doc, DocBin
+
+train_dane = "/Users/au561649/Github/DaCy/training/main/corpus/dane/train.spacy"
+dev_dane = "/Users/au561649/Github/DaCy/training/main/corpus/dane/dev.spacy"
+test_dane = "/Users/au561649/Github/DaCy/training/main/corpus/dane/test.spacy"
+
+
+nlp = spacy.blank("da")
+# train, dev, test = dane()
+train_docs = list(DocBin().from_disk(train_dane).get_docs(nlp.vocab))
+dev_docs = list(DocBin().from_disk(dev_dane).get_docs(nlp.vocab))
+test_docs = list(DocBin().from_disk(test_dane).get_docs(nlp.vocab))
+
+Doc.set_extension("split", default=None)
+
+for split, nam in zip([train_docs, dev_docs, test_docs], ["train", "dev", "test"]):
+    for doc in split:
+        doc._.split = nam
+
+# text2doc = {}
+# n_duplicates = 0  # all looks like non-actual duplicates (e.g. "stk. 2")
+# for i, doc in enumerate(test_docs + train_docs + dev_docs):
+#     if doc.text in text2doc:
+#         print(f"Duplicate found: {doc.text}")
+#         print("split:": doc._.split)
+#         n_duplicates += 1
+#     text2doc[doc.text] = doc
+
+
+# load daneplus
+path_to_data = "/Users/au561649/Github/DaCy/training/dane_plus/train.spacy"
+train_data = DocBin().from_disk(path_to_data)
+daneplus_docs = list(train_data.get_docs(nlp.vocab))
+
+text2doc = {}
+n_duplicates = 0  # No duplicates (prodigy removed them - this will be problematic when reconstructing the documents - so therefore we re-annotate the dane documents)
+for i, doc in enumerate(daneplus_docs):
+    if doc.text in text2doc:
+        print(f"Duplicate found: {doc.text}")
+        n_duplicates += 1
+    text2doc[doc.text] = doc
+
+# Add the daneplus annotations to the dane documents
+docs_to_fix = []
+for doc in train_docs + dev_docs + test_docs:
+    if doc.text in text2doc:
+        _ents_to_add = text2doc[doc.text].ents
+        ents_to_add = []
+        for ent in _ents_to_add:
+            char_span = doc.char_span(ent.start_char, ent.end_char, label=ent.label_)
+            if char_span is None:
+                print(f"Entity could not be added: {ent.text}")
+                docs_to_fix.append((doc, ent))
+                continue
+            ents_to_add.append(char_span)
+        doc.ents = ents_to_add  # type: ignore
+
+# manual fixes (due to difference in tokenization)
+doc, ent = docs_to_fix[0]
+ents = list(doc.ents)
+_ent = doc[-2:-1]
+new_ent = doc.char_span(_ent.start_char, _ent.end_char, label=ent.label_)
+print("added", new_ent, "to", doc.text)
+ents.append(new_ent)
+doc.ents = ents
+
+
+doc, ent = docs_to_fix[1]
+ents = list(doc.ents)
+_ent = doc[-3:-1]
+new_ent = doc.char_span(_ent.start_char, _ent.end_char, label=ent.label_)
+ents.append(new_ent)
+doc.ents = ents
+print("added", new_ent, "to", doc.text)
+
+doc, ent = docs_to_fix[2]
+ents = list(doc.ents)
+_ent = doc[-3:-1]
+new_ent = doc.char_span(_ent.start_char, _ent.end_char, label=ent.label_)
+ents.append(new_ent)
+doc.ents = ents
+print("added", new_ent, "to", doc.text)
+
+doc, ent = docs_to_fix[3]
+ents = list(doc.ents)
+_ent = doc[-3:-1]
+new_ent = doc.char_span(_ent.start_char, _ent.end_char, label=ent.label_)
+ents.append(new_ent)
+doc.ents = ents
+print("added", new_ent, "to", doc.text)
+
+doc, ent = docs_to_fix[4]
+ents = list(doc.ents)
+_ent = doc[-3:-1]
+new_ent = doc.char_span(_ent.start_char, _ent.end_char, label=ent.label_)
+ents.append(new_ent)
+doc.ents = ents
+print("added", new_ent, "to", doc.text)
+
+doc, ent = docs_to_fix[5]
+ents = list(doc.ents)
+_ent = doc[-3:-1]
+new_ent = doc.char_span(_ent.start_char, _ent.end_char, label=ent.label_)
+ents.append(new_ent)
+doc.ents = ents
+print("added", new_ent, "to", doc.text)
+
+
+# Save the new documents
+new_train = DocBin(docs=train_docs)
+new_dev = DocBin(docs=dev_docs)
+new_test = DocBin(docs=test_docs)
+
+new_train.to_disk("train.spacy")
+new_dev.to_disk("dev.spacy")
+new_test.to_disk("test.spacy")

dataset_creation_scripts/upload_to_hf.py

@@ -0,0 +1,29 @@
+from datasets import Dataset, DatasetDict
+from spacy.tokens import DocBin
+import spacy
+
+
+def convert_spacy_docs_to_hf_entry(doc) -> dict:
+    return doc.to_json()
+
+train = "/Users/au561649/Github/DaCy/training/dane_plus/train.spacy"
+dev = "/Users/au561649/Github/DaCy/training/dane_plus/dev.spacy"
+test = "/Users/au561649/Github/DaCy/training/dane_plus/test.spacy"
+
+nlp = spacy.blank("da")
+
+train_docs = list(DocBin().from_disk(train).get_docs(nlp.vocab))
+dev_docs = list(DocBin().from_disk(dev).get_docs(nlp.vocab))
+test_docs = list(DocBin().from_disk(test).get_docs(nlp.vocab))
+
+
+# my_list = [{"a": 1}, {"a": 2}, {"a": 3}]
+# dataset = Dataset.from_list(my_list)
+
+train_dataset = Dataset.from_list([convert_spacy_docs_to_hf_entry(doc) for doc in train_docs])
+dev_dataset = Dataset.from_list([convert_spacy_docs_to_hf_entry(doc) for doc in dev_docs])
+test_dataset = Dataset.from_list([convert_spacy_docs_to_hf_entry(doc) for doc in test_docs])
+
+dataset_dict = DatasetDict({"train": train_dataset, "dev": dev_dataset, "test": test_dataset})
+
+dataset_dict.push_to_hub("dane_plus")