Token Classification
Transformers
ONNX
Safetensors
English
Irish
distilbert
pii
de-identification
ireland
irish
gaelic
diffusion-style
denoising
ppsn
eircode
int8
dynamic-quantization
cpu
Eval Results (legacy)
Instructions to use temsa/IrishCore-DiffMask-135M-v1-rc2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use temsa/IrishCore-DiffMask-135M-v1-rc2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="temsa/IrishCore-DiffMask-135M-v1-rc2")# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("temsa/IrishCore-DiffMask-135M-v1-rc2") model = AutoModel.from_pretrained("temsa/IrishCore-DiffMask-135M-v1-rc2") - Notebooks
- Google Colab
- Kaggle
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -342,7 +342,7 @@ Additional local synthetic hardening and replay sets used during checkpoint sele
|
|
| 342 |
<!-- portfolio-comparison:start -->
|
| 343 |
## Portfolio Comparison
|
| 344 |
|
| 345 |
-
Updated: `2026-03-
|
| 346 |
|
| 347 |
Use this section for the fastest public comparison across the `temsa` PII masking portfolio.
|
| 348 |
|
|
@@ -357,6 +357,7 @@ Use this section for the fastest public comparison across the `temsa` PII maskin
|
|
| 357 |
|
| 358 |
| Repo | Stack | Full Core F1 | Q8 Core F1 | Q8 Multilingual PPSN F1 | Q8 Core ex/s |
|
| 359 |
|---|---|---:|---:|---:|---:|
|
|
|
|
| 360 |
| [`temsa/IrishCore-GlobalPointer-135M-v1-rc2`](https://huggingface.co/temsa/IrishCore-GlobalPointer-135M-v1-rc2) | GlobalPointer raw-only span-matrix | 0.9934 | 0.9934 | 0.9326 | 231.2 |
|
| 361 |
| [`temsa/OpenMed-mLiteClinical-IrishCorePII-135M-v2-rc8`](https://huggingface.co/temsa/OpenMed-mLiteClinical-IrishCorePII-135M-v2-rc8) | Raw-only token-span | 0.9737 | 0.9737 | 0.9176 | 46.1 |
|
| 362 |
| [`temsa/OpenMed-mLiteClinical-IrishCorePII-135M-v2-rc7`](https://huggingface.co/temsa/OpenMed-mLiteClinical-IrishCorePII-135M-v2-rc7) | Hybrid classifier + generated scanner spec | 1.0000 | 0.9934 | 1.0000 | 30.0 |
|
|
@@ -399,5 +400,5 @@ Finance-boundary q8 F1 is `1.0000` for `OpenMed-mLiteClinical-IrishCorePII-135M-
|
|
| 399 |
| [`temsa/OpenMed-PPSN-v5`](https://huggingface.co/temsa/OpenMed-PPSN-v5) | irish_reg_v5 raw 0.8235; irish_reg_v5 hybrid strict 1.0000 | Hybrid PPSN-only checkpoint; predates the canonical multilingual suite packaging. |
|
| 400 |
| [`temsa/OpenMed-PPSN-v4`](https://huggingface.co/temsa/OpenMed-PPSN-v4) | synthetic non-PPSN drift check only | Predates the current PPSN eval suite; no packaged apples-to-apples multilingual CPU row. |
|
| 401 |
|
| 402 |
-
If you need the strongest current raw-only Irish core model, start with `IrishCore-GlobalPointer-135M-v1-
|
| 403 |
<!-- portfolio-comparison:end -->
|
|
|
|
| 342 |
<!-- portfolio-comparison:start -->
|
| 343 |
## Portfolio Comparison
|
| 344 |
|
| 345 |
+
Updated: `2026-03-15`.
|
| 346 |
|
| 347 |
Use this section for the fastest public comparison across the `temsa` PII masking portfolio.
|
| 348 |
|
|
|
|
| 357 |
|
| 358 |
| Repo | Stack | Full Core F1 | Q8 Core F1 | Q8 Multilingual PPSN F1 | Q8 Core ex/s |
|
| 359 |
|---|---|---:|---:|---:|---:|
|
| 360 |
+
| [`temsa/IrishCore-GlobalPointer-135M-v1-rc3`](https://huggingface.co/temsa/IrishCore-GlobalPointer-135M-v1-rc3) | GlobalPointer raw-only span-matrix | 1.0000 | 1.0000 | 0.9213 | 204.9 |
|
| 361 |
| [`temsa/IrishCore-GlobalPointer-135M-v1-rc2`](https://huggingface.co/temsa/IrishCore-GlobalPointer-135M-v1-rc2) | GlobalPointer raw-only span-matrix | 0.9934 | 0.9934 | 0.9326 | 231.2 |
|
| 362 |
| [`temsa/OpenMed-mLiteClinical-IrishCorePII-135M-v2-rc8`](https://huggingface.co/temsa/OpenMed-mLiteClinical-IrishCorePII-135M-v2-rc8) | Raw-only token-span | 0.9737 | 0.9737 | 0.9176 | 46.1 |
|
| 363 |
| [`temsa/OpenMed-mLiteClinical-IrishCorePII-135M-v2-rc7`](https://huggingface.co/temsa/OpenMed-mLiteClinical-IrishCorePII-135M-v2-rc7) | Hybrid classifier + generated scanner spec | 1.0000 | 0.9934 | 1.0000 | 30.0 |
|
|
|
|
| 400 |
| [`temsa/OpenMed-PPSN-v5`](https://huggingface.co/temsa/OpenMed-PPSN-v5) | irish_reg_v5 raw 0.8235; irish_reg_v5 hybrid strict 1.0000 | Hybrid PPSN-only checkpoint; predates the canonical multilingual suite packaging. |
|
| 401 |
| [`temsa/OpenMed-PPSN-v4`](https://huggingface.co/temsa/OpenMed-PPSN-v4) | synthetic non-PPSN drift check only | Predates the current PPSN eval suite; no packaged apples-to-apples multilingual CPU row. |
|
| 402 |
|
| 403 |
+
If you need the strongest current raw-only Irish core model, start with `IrishCore-GlobalPointer-135M-v1-rc3`. If you need the fastest CPU-first raw-only line, compare it against `IrishCore-DiffMask-135M-v1-rc6`. If you need a PPSN-only artifact, compare the canonical `fp32`, `fp16`, and `q8` variants of `OpenMed-mLiteClinical-IrishPPSN-135M-v1` directly in the table above.
|
| 404 |
<!-- portfolio-comparison:end -->
|