desert-ant-labs
/

redact

Token Classification

Model card Files Files and versions

laurier-rochon commited on about 8 hours ago

Commit

50dd2c5

·

verified ·

1 Parent(s): fa5108b

Coverage reads 27 languages

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -45,7 +45,7 @@ pipeline_tag: token-classification
 # redact: on-device multilingual PII redaction
 Detects and redacts personal data (names, addresses, emails, phone numbers,
-cards, IBANs, national IDs and more) in text across **24 EU languages and more**
 (see [Languages](#languages)). A BIOES token classifier plus a portable, dependency-free
 deterministic layer for structured IDs. The deployable model is **~11.6 MB**
 (4-bit Core ML on Apple) or **~24.5 MB** (int8 LiteRT `.tflite` on Android, Linux
@@ -129,7 +129,7 @@ covers one of the 27 languages listed below.
 ## Languages
-**27 languages**: all 24 official EU languages, plus 3 more.
 Latin, Greek and Cyrillic scripts.
 ### The 24 EU languages

 # redact: on-device multilingual PII redaction
 Detects and redacts personal data (names, addresses, emails, phone numbers,
+cards, IBANs, national IDs and more) in text across **27 languages**
 (see [Languages](#languages)). A BIOES token classifier plus a portable, dependency-free
 deterministic layer for structured IDs. The deployable model is **~11.6 MB**
 (4-bit Core ML on Apple) or **~24.5 MB** (int8 LiteRT `.tflite` on Android, Linux
 ## Languages
+**27 languages**: every official EU language, plus 3 more.
 Latin, Greek and Cyrillic scripts.
 ### The 24 EU languages