fix readme

nell.py

@@ -5,7 +5,7 @@ import datasets
 logger = datasets.logging.get_logger(__name__)
 _DESCRIPTION = """Few shots link prediction dataset. """
 _NAME = "nell"
-_VERSION = "0.0.7"
+_VERSION = "0.0.8"
 _CITATION = """
 @inproceedings{xiong-etal-2018-one,
     title = "One-Shot Relational Learning for Knowledge Graphs",
@@ -76,10 +76,8 @@ class NELL(datasets.GeneratorBasedBuilder):
                 {
                     "relation": datasets.Value("string"),
                     "head": datasets.Value("string"),
-                    "head_entity": datasets.Value("string"),
                     "head_type": datasets.Value("string"),
                     "tail": datasets.Value("string"),
-                    "tail_entity": datasets.Value("string"),
                     "tail_type": datasets.Value("string"),
                 }
             ),

stats.py

@@ -1,70 +1,13 @@
-import pandas as pd
-from datasets import load_dataset
 from itertools import chain
 
-tmp = {}
-for _type in ['nell_filter']:
-    with open(f"data/{_type}.vocab.txt") as f:
-        vocab = pd.DataFrame([i.split("\t") for i in f.read().split('\n') if len(i) > 0], columns=["entity", "type"])
-    vocab_df = vocab.groupby("type").count().sort_values(by="entity", ascending=False)
-    vocab_df['sample'] = [
-        ", ".join(vocab[vocab.type == i].sample(min(3, sum(vocab.type == i)))["entity"].values.tolist()) for i in
-        vocab_df.index]
-    tmp[_type] = vocab_df
-keys = set(list(chain(*[list(v.index) for v in tmp.values()])))
-df = pd.DataFrame([{
-    "entity_type": k,
-    "nell": tmp["nell"].loc[k]['entity'] if k in tmp["nell"].index else 0,
-    "nell_filter": tmp["nell_filter"].loc[k]['entity'] if k in tmp["nell_filter"].index else 0,
-    "sample": tmp["nell"].loc[k]['sample'] if k in tmp["nell"].index else tmp["nell_filter"].loc[k]['sample']
-} for k in keys]).sort_values(by="nell", ascending=False)
-df = pd.concat([df, pd.DataFrame([{
-    "entity_type": "SUM",
-    'nell': df['nell'].sum(),
-    'nell_filter': df['nell_filter'].sum(),
-    'sample': ""}])])
-df.to_csv(f"stats/stats.vocab.csv", index=False)
-print(f"\nVocab Size")
-print(df.to_markdown(index=False))
-
-for split in ['train', 'validation', 'test']:
-    print(split)
+import pandas as pd
+from datasets import load_dataset
 
-    tmp = {}
-    for _type in ['nell_filter']:
-        data = load_dataset("relbert/nell", _type, split=split)
-        df = data.to_pandas()
+def get_stats(split):
+    data = load_dataset("relbert/nell", split=split)
+    df = data.to_pandas()
 
-        tail = df.groupby("tail_type")['relation'].count().to_dict()
-        head = df.groupby("head_type")['relation'].count().to_dict()
-        k = set(list(tail.keys()) + list(head.keys()))
-        df_types = pd.DataFrame([{"entity_type": _k, "tail": tail[_k] if _k in tail else 0, "head": head[_k] if _k in head else 0} for _k in k])
-        df_types.index = df_types.pop("entity_type")
-        tmp[_type] = df_types
 
-    keys = set(list(chain(*[list(v.index) for v in tmp.values()])))
-    df = pd.DataFrame([{
-        "entity_type": k,
-        "nell (head)": tmp["nell"].loc[k]['head'] if k in tmp["nell"].index else 0,
-        "nell_filter (head)": tmp["nell_filter"].loc[k]['head'] if k in tmp["nell_filter"].index else 0,
-        "nell (tail)": tmp["nell"].loc[k]['tail'] if k in tmp["nell"].index else 0,
-        "nell_filter (tail)": tmp["nell_filter"].loc[k]['tail'] if k in tmp["nell_filter"].index else 0,
-    } for k in keys]).sort_values(by="nell (head)", ascending=False)
-    df.to_csv(f"stats/stats.{split}.entity.csv", index=False)
-    print(f"\nHead/Tail Size")
-    print(df.to_markdown(index=False))
+s = 'test'
 
-    tmp = {}
-    data = load_dataset("relbert/nell", split=split)
-    df = data.to_pandas()
-    tmp[_type] = df.groupby("relation").count()['head']
-    keys = set(list(chain(*[list(v.index) for v in tmp.values()])))
-    df = pd.DataFrame([{
-        "relation_type": k,
-        "nell": tmp["nell"].loc[k] if k in tmp["nell"].index else 0,
-        "nell_filter": tmp["nell_filter"].loc[k] if k in tmp["nell_filter"].index else 0,
-    } for k in keys]).sort_values(by="nell", ascending=False)
-    df.to_csv(f"stats/stats.{split}.relation.csv", index=False)
-    print(f"\nRelation Size")
-    print(df.to_markdown(index=False))