switch to dask only with proper partition size

README.md

@@ -21,7 +21,6 @@ configs:
 - config_name: question-answer
   data_files: 
   - split: test
-    path: "data/test.parquet"
+    path: "data/test.parquet/*"
 ---
-
-Derives from https://www.kaggle.com/datasets/rtatman/questionanswer-dataset?resource=download we generated our own subset using `generate.py`.
+Derives from https://www.kaggle.com/datasets/rtatman/questionanswer-dataset?resource=download we generated our own subset using `generate.py`.

data/passages.parquet/part.0.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b94475e002695ada63d56ecc7a769e36b40cd565814600a0d8cb09d31c3370b8
-size 202740
+oid sha256:87733ad9242f56a08396e9a00e0551a1539f286ecf032d24ecff34dba4689a67
+size 797357

data/passages.parquet/part.1.parquet

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:2d6856418a9a52e72173e5f76001a8cae475a019ccf480b94729c55080f1ff71
-size 200292

data/passages.parquet/part.2.parquet

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:418ce71d54decd51e70ef756f8a5f01df6d7f6bde61d2c64911d1dd1878ca73e
-size 193067

data/passages.parquet/part.3.parquet

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:639fe3406b2a757d0ff0f0b534623f17cb7ae7ae14b544cd8b09cbb506ba22df
-size 211512

data/{test.parquet → test.parquet/part.0.parquet}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bde1b058cd5d9f19afb6af7a78f5e943e8651a4fc1e7e821c5a672909530ef4c
+oid sha256:8480d74e65ed193879cfe4549d4e9239cfef6dce0e0e82bde1f3a1a220837641
 size 54366

generate.py

@@ -14,7 +14,7 @@ df = df.drop_duplicates(subset="question")
 df.index.name = "id"
 
 path_to_file = "./data/test.parquet"
-df.to_parquet(path_to_file)
+dd.from_pandas(df, npartitions=1).to_parquet(path_to_file)
 
 path_to_file = "./raw_data/text_data/S08*clean"
 
@@ -28,4 +28,4 @@ passages = passages.dropna()
 passages.index.name = "id"
 
 path_to_file = "./data/passages.parquet"
-dd.from_pandas(passages, npartitions=4).to_parquet(path_to_file)
+dd.from_pandas(passages, npartitions=1).to_parquet(path_to_file)