Add model card

README.md

@@ -1,130 +1,61 @@
 ---
+language: en
+license: apache-2.0
 tags:
-- setfit
-- sentence-transformers
-- text-classification
-- generated_from_setfit_trainer
-widget: []
-metrics:
-- accuracy
+  - setfit
+  - text-classification
+  - dlp
+  - security
+  - pii-detection
+  - data-loss-prevention
 pipeline_tag: text-classification
-library_name: setfit
-inference: true
 ---
 
-# SetFit
+# spidercob/dlp-intent-classifier
 
-This is a [SetFit](https://github.com/huggingface/setfit) model that can be used for Text Classification. A [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance is used for classification.
+A SetFit model trained to classify DLP (Data Loss Prevention) regex match findings as real sensitive data vs. false positives.
 
-The model has been trained using an efficient few-shot learning technique that involves:
+## Labels
 
-1. Fine-tuning a [Sentence Transformer](https://www.sbert.net) with contrastive learning.
-2. Training a classification head with features from the fine-tuned Sentence Transformer.
+| Label | Description | Default Action |
+|---|---|---|
+| `REAL_DATA` | Genuine PII, credential, or secret | BLOCK (if conf > 0.7) |
+| `TEST_DATA` | Test/mock/example value | ALLOW (if conf > 0.8) |
+| `DOCUMENTATION` | Regex hit inside a doc or code comment | ALLOW |
+| `NOISE` | False positive or low-signal hit | IGNORE |
 
-## Model Details
-
-### Model Description
-- **Model Type:** SetFit
-<!-- - **Sentence Transformer:** [Unknown](https://huggingface.co/unknown) -->
-- **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
-- **Maximum Sequence Length:** 256 tokens
-- **Number of Classes:** 4 classes
-<!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
-<!-- - **Language:** Unknown -->
-<!-- - **License:** Unknown -->
-
-### Model Sources
-
-- **Repository:** [SetFit on GitHub](https://github.com/huggingface/setfit)
-- **Paper:** [Efficient Few-Shot Learning Without Prompts](https://arxiv.org/abs/2209.11055)
-- **Blogpost:** [SetFit: Efficient Few-Shot Learning Without Prompts](https://huggingface.co/blog/setfit)
-
-## Uses
-
-### Direct Use for Inference
-
-First install the SetFit library:
-
-```bash
-pip install setfit
-```
-
-Then you can load this model and run inference.
+## Usage
 
 ```python
 from setfit import SetFitModel
 
-# Download from the 🤗 Hub
-model = SetFitModel.from_pretrained("setfit_model_id")
-# Run inference
-preds = model("I loved the spiderman movie!")
-```
-
-<!--
-### Downstream Use
+model = SetFitModel.from_pretrained("spidercob/dlp-intent-classifier")
 
-*List how someone could finetune this model on their own dataset.*
--->
+text = "Context surrounding a ssn: Patient record: Maria Garcia SSN=523-89-4521"
+prediction = model.predict([text])
+# ['REAL_DATA']
 
-<!--
-### Out-of-Scope Use
-
-*List how the model may foreseeably be misused and address what users ought not to do with the model.*
--->
-
-<!--
-## Bias, Risks and Limitations
-
-*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
--->
-
-<!--
-### Recommendations
-
-*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
--->
-
-## Training Details
-
-### Framework Versions
-- Python: 3.12.12
-- SetFit: 1.1.3
-- Sentence Transformers: 5.2.2
-- Transformers: 4.57.6
-- PyTorch: 2.10.0
-- Datasets: 5.0.0
-- Tokenizers: 0.22.2
+probs = model.predict_proba([text])
+# [[0.02, 0.01, 0.01, 0.96]]
+```
 
-## Citation
+## Input format
 
-### BibTeX
-```bibtex
-@article{https://doi.org/10.48550/arxiv.2209.11055,
-    doi = {10.48550/ARXIV.2209.11055},
-    url = {https://arxiv.org/abs/2209.11055},
-    author = {Tunstall, Lewis and Reimers, Nils and Jo, Unso Eun Seo and Bates, Luke and Korat, Daniel and Wasserblat, Moshe and Pereg, Oren},
-    keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
-    title = {Efficient Few-Shot Learning Without Prompts},
-    publisher = {arXiv},
-    year = {2022},
-    copyright = {Creative Commons Attribution 4.0 International}
-}
+```
+Context surrounding a {finding_type}: {context_text}
 ```
 
-<!--
-## Glossary
-
-*Clearly define terms in order to be accessible across audiences.*
--->
+- `finding_type`: the DLP pattern that matched (e.g. `ssn`, `credit_card`, `aws_access_key`, `email`, `api_key`)
+- `context_text`: surrounding content snippet, up to ~400 characters
 
-<!--
-## Model Card Authors
+## Training
 
-*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
--->
+- **Base model**: `sentence-transformers/all-MiniLM-L6-v2`
+- **Method**: SetFit contrastive fine-tuning
+- **Training examples**: 256 (production DLP findings + curated synthetic examples)
+- **Test accuracy**: 90.4%
+- **Training time**: ~8 minutes on Apple M-series
 
-<!--
-## Model Card Contact
+## Fine-tune on your own data
 
-*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
--->
+See [github.com/spidercob/dlp-intent-classifier](https://github.com/spidercob/dlp-intent-classifier) for the full training pipeline.