broadfield-dev commited on
Commit
2d09f50
·
verified ·
1 Parent(s): caa2485

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +48 -8
README.md CHANGED
@@ -1,12 +1,52 @@
1
  ---
2
- title: Privacy Filter Demo
3
- emoji: 🔥
4
- colorFrom: indigo
5
- colorTo: yellow
6
- sdk: gradio
7
- sdk_version: 6.13.0
8
- app_file: app.py
9
  pinned: false
 
 
10
  ---
11
 
12
- Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ title: Privacy Filter Flask Demo
3
+ emoji: 🔒
4
+ colorFrom: blue
5
+ colorTo: purple
6
+ sdk: docker
 
 
7
  pinned: false
8
+ license: apache-2.0
9
+ short_description: Flask demo for OpenAI Privacy Filter PII detection model
10
  ---
11
 
12
+ # 🔒 OpenAI Privacy Filter Flask Demo
13
+
14
+ A Flask-based web interface to demonstrate OpenAI's Privacy Filter model for detecting and masking Personally Identifiable Information (PII) in text.
15
+
16
+ ## Model
17
+
18
+ **openai/privacy-filter** - A bidirectional token-classification model trained specifically for PII detection.
19
+
20
+ ## Detects 8 Privacy Categories
21
+
22
+ | Category | Description | Example |
23
+ |----------|-------------|---------|
24
+ | `private_person` | Names of individuals | "Alice Smith", "John Doe" |
25
+ | `private_email` | Email addresses | "alice@example.com" |
26
+ | `private_phone` | Phone numbers | "(555) 123-4567" |
27
+ | `private_address` | Physical addresses | "123 Main Street, NYC" |
28
+ | `account_number` | Account/ID numbers | "Acct: 12345678" |
29
+ | `secret` | Secrets/credentials | Passwords, API keys, tokens |
30
+ | `private_url` | Personal URLs | Private links |
31
+ | `private_date` | Personal dates | Birthdays, dates of birth |
32
+
33
+ ## Usage
34
+
35
+ Enter any text containing PII in the web interface and the model will:
36
+ 1. **Highlight** detected PII entities with color coding
37
+ 2. **Label** each entity by type
38
+ 3. **Show confidence scores** for each detection
39
+
40
+ ## Technical Details
41
+
42
+ - **Framework**: Flask
43
+ - **Model**: openai/privacy-filter (1.5B params, 50M active)
44
+ - **License**: Apache 2.0
45
+ - **Context Window**: 128,000 tokens
46
+ - **Architecture**: Bidirectional token classifier with Banded Attention
47
+
48
+ ## Links
49
+
50
+ - [Model Card](https://huggingface.co/openai/privacy-filter)
51
+ - [OpenAI Privacy Filter GitHub](https://github.com/openai/privacy-filter)
52
+ - [Model Paper/Card PDF](https://cdn.openai.com/pdf/c66281ed-b638-456a-8ce1-97e9f5264a90/OpenAI-Privacy-Filter-Model-Card.pdf)