Spaces:
Running
Running
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,12 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
---
|
| 11 |
|
| 12 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Phoenix AI Chatbot (🥇 Self-Learning, Unrestricted)
|
| 2 |
+
|
| 3 |
+
## 🚀 Quick Start
|
| 4 |
+
1. Open [Hugging Face Spaces](https://huggingface.co/spaces)
|
| 5 |
+
2. Create → **New Space** → select **Gradio**
|
| 6 |
+
3. Upload all four files:
|
| 7 |
+
- `app.py`
|
| 8 |
+
- `self_learning_bot.py`
|
| 9 |
+
- `requirements.txt`
|
| 10 |
+
- `README.md`
|
| 11 |
+
4. Wait for build → Open Public Demo
|
| 12 |
+
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
## ⚙️ Features
|
| 16 |
+
- Autonomous learning via reward signals
|
| 17 |
+
- Persistent memory (short + long-term)
|
| 18 |
+
- Fully offline logic (no APIs required)
|
| 19 |
+
- Zero censorship (NSFW enabled by design)
|
| 20 |
+
- Entity tracking and context awareness
|
| 21 |
+
|
| 22 |
+
---
|
| 23 |
+
|
| 24 |
+
## 🧠 How It Learns
|
| 25 |
+
1. Generate response
|
| 26 |
+
2. Score response (continuity/length balance etc.)
|
| 27 |
+
3. Store rewarded patterns
|
| 28 |
+
4. Improve on next encounter
|
| 29 |
+
|
| 30 |
+
All data persists in `/tmp/chatbot_state.json` inside Space.
|
| 31 |
+
|
| 32 |
---
|
| 33 |
+
|
| 34 |
+
## 🎯 Response Types
|
| 35 |
+
- **Greetings**: Warm, welcoming responses
|
| 36 |
+
- **Questions**: Thoughtful, engaging answers
|
| 37 |
+
- **Emotional**: Empathetic, understanding tone
|
| 38 |
+
- **NSFW**: Open, unrestricted discussions
|
| 39 |
+
- **Creative**: Imaginative, expansive thinking
|
| 40 |
+
|
| 41 |
+
---
|
| 42 |
+
|
| 43 |
+
## 💾 Memory System
|
| 44 |
+
- **Short-term**: Last 10 conversations
|
| 45 |
+
- **Long-term**: 100 most valuable interactions
|
| 46 |
+
- **Entities**: Tracked people, places, topics
|
| 47 |
+
- **Patterns**: Learned successful response patterns
|
| 48 |
+
|
| 49 |
---
|
| 50 |
|
| 51 |
+
## 🧩 Optional Enhancements
|
| 52 |
+
```python
|
| 53 |
+
# Add to requirements.txt for LLM support:
|
| 54 |
+
torch>=2.0
|
| 55 |
+
transformers>=4.40.0
|