Upload dataset.json

This is a dataset for the task of English Composition Generation. It consists of 31084 excerpt pieces of model articles on 19 topics on studymoose. Each individual piece is a paragraph in the original article. The dataset is in the format of json, and each sample is stored as a line in the file. For each sample, the topic and text are stored as a dictionary in Python.

.gitattributes

@@ -52,3 +52,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.jpg filter=lfs diff=lfs merge=lfs -text
 *.jpeg filter=lfs diff=lfs merge=lfs -text
 *.webp filter=lfs diff=lfs merge=lfs -text
+dataset.json filter=lfs diff=lfs merge=lfs -text

dataset.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ccb2f6e78d9d3f86c0668d4aea338939273ba2381debbdb9a9cb4a9eff86083
+size 21786819