add _preprended

README.md

@@ -22,11 +22,14 @@ This dataset is compatible with the [`run_summarization.py`](https://github.com/
 
 # Configs
 4 possibles configs:
-- `roberta` will concatenate documents with "\</s\>" (default)
+- `roberta` will concatenate documents with "\</s\>"
 - `newline` will concatenate documents with "\n"
 - `bert` will concatenate documents with "[SEP]"
 - `list` will return the list of documents instead of a single string
 
+Add `_prepended` to config name to prepend the speaker name before each dialogue: `speaker: text` \
+Default is `roberta_prepended` (compatible with BART).
+
 ### Data Fields
 
 - `id`: paper id

mediasum.py

@@ -62,9 +62,29 @@ class MediaSumSummarizationDataset(datasets.GeneratorBasedBuilder):
             version=datasets.Version("1.0.0"),
             description="MediaSum dataset for summarization, document",
         ),
+        MediaSumSummarizationConfig(
+            name="newline_prepended",
+            version=datasets.Version("1.0.0"),
+            description="MediaSum dataset for summarization, concat sections",
+        ),
+        MediaSumSummarizationConfig(
+            name="roberta_prepended",
+            version=datasets.Version("1.0.0"),
+            description="MediaSum dataset for summarization, document",
+        ),
+        MediaSumSummarizationConfig(
+            name="bert_prepended",
+            version=datasets.Version("1.0.0"),
+            description="MediaSum dataset for summarization, document",
+        ),
+        MediaSumSummarizationConfig(
+            name="list_prepended",
+            version=datasets.Version("1.0.0"),
+            description="MediaSum dataset for summarization, document",
+        ),
     ]
 
-    DEFAULT_CONFIG_NAME = "roberta"
+    DEFAULT_CONFIG_NAME = "roberta_prepended"
 
     def _info(self):
         # Should return a datasets.DatasetInfo object
@@ -102,12 +122,12 @@ class MediaSumSummarizationDataset(datasets.GeneratorBasedBuilder):
     
     def _generate_examples(self, filepath):
         """Generate MediaSumSummarization examples."""
-        if self.config.name == "newline":
+        if "newline" in self.config.name:
             join_ = "\n"
-        elif self.config.name == "roberta":
+        elif "roberta" in self.config.name:
             join_ = "</s>"
-        elif self.config.name == "bert": 
-            join_ = "[SEP]"
+        elif "bert" in self.config.name: 
+            join_ = " [SEP] "
 
         with open(filepath, encoding="utf-8") as f:
             for id_, row in enumerate(f):
@@ -118,9 +138,13 @@ class MediaSumSummarizationDataset(datasets.GeneratorBasedBuilder):
                 'document': List[str],
                 """
                 
-                document = data["utt"]
+                documents = data["utt"]
+
+                if "_prepended" in self.config.name:
+                    names = data["speaker"]
+                    documents = [name + ": " + document for name, document in zip(names, documents)]
                 
                 if self.config.name != "list":
-                    document = join_.join(document)
+                    documents = join_.join(documents)
                 summary = data["summary"]
-                yield id_, {"document": document, "summary": summary}
+                yield id_, {"document": documents, "summary": summary}