Token Classification
Transformers
ONNX
Safetensors
English
distilbert
pii
privacy
experimental
datafog
Instructions to use DataFog/pii-en-65m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use DataFog/pii-en-65m with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="DataFog/pii-en-65m")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("DataFog/pii-en-65m") model = AutoModelForTokenClassification.from_pretrained("DataFog/pii-en-65m", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download bundle/rust-parity.json from DataFog/pii-en-65m: direct link, hf CLI and curl.
- Browser
- Download file 960 Bytes
-
https://huggingface.co/DataFog/pii-en-65m/resolve/main/bundle/rust-parity.json
- Command line
-
hf download hf://DataFog/pii-en-65m/bundle/rust-parity.json
-
curl -L -o rust-parity.json https://huggingface.co/DataFog/pii-en-65m/resolve/main/bundle/rust-parity.json
960 Bytes
| { | |
| "bundle": "artifacts/bundle", | |
| "cases": [ | |
| { | |
| "bytes": 0, | |
| "windows": 1, | |
| "findings": 0 | |
| }, | |
| { | |
| "bytes": 11, | |
| "windows": 1, | |
| "findings": 2 | |
| }, | |
| { | |
| "bytes": 52, | |
| "windows": 1, | |
| "findings": 2 | |
| }, | |
| { | |
| "bytes": 26, | |
| "windows": 1, | |
| "findings": 0 | |
| }, | |
| { | |
| "bytes": 28, | |
| "windows": 1, | |
| "findings": 3 | |
| }, | |
| { | |
| "bytes": 9933, | |
| "windows": 5, | |
| "findings": 3 | |
| }, | |
| { | |
| "bytes": 4559, | |
| "windows": 3, | |
| "findings": 2 | |
| } | |
| ], | |
| "max_abs_logit_difference": 0.0, | |
| "logits_bit_identical": true, | |
| "token_ids_masks_global_byte_offsets_parity": true, | |
| "findings_redaction_parity": true, | |
| "actual_python_predictor_parity": true, | |
| "malformed_request_explicit_error_and_recovery": true, | |
| "cold_process_debug_batch_seconds": 1.1464320829836652, | |
| "note": "Parity fixtures prove implementation consistency, not detection quality." | |
| } | |