modify app.py

app.py

@@ -10,6 +10,21 @@ from pathlib import Path
 # 定义 tiktoken 编码器
 encoding = tiktoken.get_encoding("cl100k_base")
 
+_CITATION = """\
+@InProceedings{huggingface:dataset,
+title = {MGT detection},
+author={Trustworthy AI Lab},
+year={2024}
+}
+"""
+
+_DESCRIPTION = """\
+For detecting machine generated text.
+"""
+
+_HOMEPAGE = ""
+_LICENSE = ""
+
 # MGTHuman 类
 class MGTHuman(datasets.GeneratorBasedBuilder):
     VERSION = datasets.Version("1.0.0")
@@ -23,6 +38,22 @@ class MGTHuman(datasets.GeneratorBasedBuilder):
     ]
     DEFAULT_CONFIG_NAME = "human"
 
+    def _info(self):
+        features = datasets.Features(
+            {
+                "id": datasets.Value("int32"),
+                "text": datasets.Value("string"),
+                "file": datasets.Value("string"),
+            }
+        )
+        return datasets.DatasetInfo(
+            description=_DESCRIPTION,
+            features=features,
+            homepage=_HOMEPAGE,
+            license=_LICENSE,
+            citation=_CITATION,
+        )
+
     def truncate_text(self, text, max_tokens=2048):
         tokens = encoding.encode(text, allowed_special={'<|endoftext|>'})
         if len(tokens) > max_tokens: