Datasets:

Finnish-NLP
/

mc4_fi_cleaned

Tasks:

Text Generation

Sub-tasks:

language-modeling

masked-language-modeling

Languages:

Multilinguality:

Size Categories:

Source Datasets:

Tags:

Dataset card Viewer Files Files and versions Community

aapot commited on Nov 18, 2021

Commit

1c59d31

•

1 Parent(s): 05759f3

Remove unnecessary dataset columns

Files changed (3) hide show

clean_data.py +2 -0
train.csv +2 -2
valid.csv +2 -2

clean_data.py CHANGED Viewed

@@ -54,5 +54,7 @@ num_rows = fi_val_cleaned.num_rows
 print(f"Final cleaned dataset val rows {num_rows}")
 # SAVE TO DISK
 fi_train_cleaned.to_csv("train.csv", num_proc=64)
 fi_val_cleaned.to_csv("valid.csv", num_proc=64)

 print(f"Final cleaned dataset val rows {num_rows}")
 # SAVE TO DISK
+fi_train_cleaned = fi_train_cleaned.remove_columns(["alphabet_len", "number_len", "upper_len", "total_len", "predicted_lang", "predicted_lang_percentage", "alphabet_ratio", "number_ratio", "upper_ratio"])
+fi_val_cleaned = fi_val_cleaned.remove_columns(["alphabet_len", "number_len", "upper_len", "total_len", "predicted_lang", "predicted_lang_percentage", "alphabet_ratio", "number_ratio", "upper_ratio"])
 fi_train_cleaned.to_csv("train.csv", num_proc=64)
 fi_val_cleaned.to_csv("valid.csv", num_proc=64)

train.csv CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc9f4e2d48ba5fdd16e8636c17f775bcab1284958705b7cbb097005d1fa8f579
-size 66554165333

 version https://git-lfs.github.com/spec/v1
+oid sha256:0fe3af1835a6120af364f2b8a9ad6526e5d27cb0dd0ad8ef73095f33f26762b6
+size 64653060288

valid.csv CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff43367cf849f61f749cad72307c8bbdb67b46f553c004f35ad07ca683a83a9a
-size 65904851

 version https://git-lfs.github.com/spec/v1
+oid sha256:43093465ebea72484795cdd86d226390a97479e5a98f1a1aaa63dfee75b438ac
+size 64018640