Marek4321 commited on
Commit
0ec7101
·
verified ·
1 Parent(s): c3368e8

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +81 -7
README.md CHANGED
@@ -1,11 +1,85 @@
1
  ---
2
- title: Prompt Tester
3
- emoji: 📈
4
- colorFrom: purple
5
- colorTo: gray
6
- sdk: static
 
 
7
  pinned: false
8
- license: apache-2.0
9
  ---
10
 
11
- Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ title: Prompt A/B Tester
3
+ emoji: 🔬
4
+ colorFrom: blue
5
+ colorTo: purple
6
+ sdk: streamlit
7
+ sdk_version: 1.28.0
8
+ app_file: app.py
9
  pinned: false
10
+ license: mit
11
  ---
12
 
13
+ # Prompt A/B Tester
14
+
15
+ Web application for testing and comparing the performance of two OpenAI GPT prompt versions with A/B testing methodology.
16
+
17
+ ## Features
18
+
19
+ - **API Key Management**: Securely store OpenAI API key in Streamlit secrets
20
+ - **Model Selection**: Choose from available GPT models (gpt-4o, gpt-4, gpt-3.5-turbo)
21
+ - **Parameter Configuration**:
22
+ - Temperature (0.0 - 2.0)
23
+ - Max Tokens (100 - 4000)
24
+ - Number of responses (1-50)
25
+ - **Two Input Methods**:
26
+ - Upload `.md` or `.txt` files
27
+ - Paste text directly
28
+ - **Progress Tracking**: Real-time progress bars and logs
29
+ - **Interactive Rating**: Rate responses on a 1-5 scale
30
+ - **Results Visualization**: Compare prompts with statistics
31
+ - **CSV Export**: Download results with metadata
32
+
33
+ ## Usage
34
+
35
+ ### 1. Configuration
36
+
37
+ 1. Go to "⚙️ Configuration" tab
38
+ 2. Enter your OpenAI API key
39
+ 3. Select model and adjust parameters
40
+ 4. Save settings
41
+
42
+ ### 2. Run Test
43
+
44
+ 1. Go to "🚀 Test" tab
45
+ 2. Either upload two files OR paste two prompts
46
+ 3. Click "Run Test"
47
+ 4. Monitor progress
48
+
49
+ ### 3. Rate Responses
50
+
51
+ 1. After generation, go to "⭐ Rating" tab
52
+ 2. Rate each response (1-5)
53
+ 3. Click "Calculate Results"
54
+
55
+ ### 4. View Results
56
+
57
+ 1. Review statistics in "📊 Results" tab
58
+ 2. Export to CSV if needed
59
+
60
+ ## Configuration on Hugging Face Spaces
61
+
62
+ To use this app on Hugging Face Spaces:
63
+
64
+ 1. Go to Settings → Repository secrets
65
+ 2. Add a new secret:
66
+ - **Name**: `OPENAI_API_KEY`
67
+ - **Value**: Your OpenAI API key
68
+
69
+ ## Cost Estimation
70
+
71
+ Approximate costs per test (2025 pricing):
72
+
73
+ | Model | 10 responses (5A + 5B) | 20 responses (10A + 10B) |
74
+ |-------|------------------------|---------------------------|
75
+ | gpt-4o | ~$0.05 | ~$0.10 |
76
+ | gpt-4-turbo | ~$0.15 | ~$0.30 |
77
+ | gpt-4 | ~$0.30 | ~$0.60 |
78
+ | gpt-3.5-turbo | ~$0.01 | ~$0.02 |
79
+
80
+ ## Project Information
81
+
82
+ **Version**: 2.0 (Streamlit)
83
+ **Original Project**: Manuscripto - Medical Editorial Assistant
84
+ **Created**: 2025-10-14
85
+ **License**: MIT