Update README.md

README.md

@@ -6,9 +6,10 @@ Original dataset is available on the HuggingFace Hub [here](https://huggingface.
 This is the same as the `snli_format` split of the SciTail dataset available on the HuggingFace Hub (i.e., same data, same splits, etc).
 The only differences are the following:
 
-- selecting only the columns `["sentence1", "sentence2", "gold_label"]`
-- renaming columns with the following mapping `{"sentence1": "premise", "sentence2": "hypothesis", "gold_label": "label"}`
-- encoding labels with the following mapping `{"entailment": 0, "neutral": 1, "contradiction": 2}`
+- selecting only the columns `["sentence1", "sentence2", "gold_label", "label"]`
+- renaming columns with the following mapping `{"sentence1": "premise", "sentence2": "hypothesis"}`
+- creating a new column "label" from "gold_label" with the following mapping `{"entailment": "entailment", "neutral": "not_entailment"}`
+- encoding labels with the following mapping `{"not_entailment": 0, "entailment": 1}`
 
 Note that there are 10 overlapping instances (as found by merging on columns "label", "premise", and "hypothesis") between
 `train` and `test` splits.
@@ -29,16 +30,17 @@ for name, df_ in dd.items():
     df = df[["sentence1", "sentence2", "gold_label"]]
 
     # rename columns
-    df = df.rename(columns={"sentence1": "premise", "sentence2": "hypothesis", "gold_label": "label"})
+    df = df.rename(columns={"sentence1": "premise", "sentence2": "hypothesis"})
 
     # encode labels
-    df["label"] = df["label"].map({"entailment": 0, "neutral": 1, "contradiction": 2})
+    df["label"] = df["gold_label"].map({"entailment": "entailment", "neutral": "not_entailment"})
+    df["label"] = df["label"].map({"not_entailment": 0, "entailment": 1})
 
     # cast to dataset
     features = Features({
         "premise": Value(dtype="string", id=None),
         "hypothesis": Value(dtype="string", id=None),
-        "label": ClassLabel(num_classes=3, names=["entailment", "neutral", "contradiction"]),
+        "label": ClassLabel(num_classes=2, names=["not_entailment", "entailment"]),
     })
     ds[name] = Dataset.from_pandas(df, features=features)