Datasets:

rag-datasets
/

rag-mini-wikipedia

Tasks:

Question Answering

Sentence Similarity

Modalities:

Formats:

Languages:

Size:

Tags:

information-retrieval

Libraries:

License:

Dataset card Data Studio Files Files and versions Community

tillwenke commited on Oct 27, 2023

Commit

d282d8c

·

1 Parent(s): f046266

rearrange data

Files changed (6) hide show

{dataset → data}/passages.parquet/part.0.parquet +0 -0
{dataset → data}/passages.parquet/part.1.parquet +0 -0
{dataset → data}/passages.parquet/part.2.parquet +0 -0
{dataset → data}/passages.parquet/part.3.parquet +0 -0
dataset/eval.parquet → data/test.parquet +0 -0
generate.py +2 -2

{dataset → data}/passages.parquet/part.0.parquet RENAMED Viewed

File without changes

{dataset → data}/passages.parquet/part.1.parquet RENAMED Viewed

File without changes

{dataset → data}/passages.parquet/part.2.parquet RENAMED Viewed

File without changes

{dataset → data}/passages.parquet/part.3.parquet RENAMED Viewed

File without changes

dataset/eval.parquet → data/test.parquet RENAMED Viewed

File without changes

generate.py CHANGED Viewed

@@ -11,7 +11,7 @@ df.rename(columns={"Question": "question", "Answer": "answer"}, inplace=True)
 df = df[["question", "answer"]]
 df = df.dropna()
 df = df.drop_duplicates(subset="question")
-path_to_file = "./dataseeval.parquet"
 df.to_parquet(path_to_file)
@@ -27,6 +27,6 @@ passages = passages.dropna()
 passages.index = passages.index.astype(str)
-path_to_file = "./dataset/passages.parquet"
 dd.from_pandas(passages, npartitions=4).to_parquet(path_to_file)

 df = df[["question", "answer"]]
 df = df.dropna()
 df = df.drop_duplicates(subset="question")
+path_to_file = "./data/test.parquet"
 df.to_parquet(path_to_file)
 passages.index = passages.index.astype(str)
+path_to_file = "./data/passages.parquet"
 dd.from_pandas(passages, npartitions=4).to_parquet(path_to_file)