|
# rad_cs_myfirstmodel - My First AI Computer Science Specialist |
|
|
|
π **My first published AI model!** A specialized 4.7GB AI that focuses exclusively on computer science topics. |
|
|
|
## π Model Overview |
|
- **Model Name**: rad_cs_myfirstmodel |
|
- **Model ID**: 0bb1cdc93d3a |
|
- **Size**: 4.7 GB |
|
- **Base Model**: qwen2.5:7b |
|
- **Specialization**: Computer Science & Programming |
|
- **Creator**: [@sanual](https://huggingface.co/sanual) (First model!) |
|
|
|
## π― What Makes This Special |
|
This is my **first AI model creation!** It transforms a general AI into a focused computer science expert that: |
|
|
|
- β
**Only talks about CS topics** - No weather, cooking, or off-topic responses |
|
- β
**Provides technical accuracy** - Optimized for programming and algorithms |
|
- β
**Works offline** - Complete privacy, no internet required |
|
- β
**Custom trained** - Specialized system prompts for CS expertise |
|
|
|
## π» Installation & Usage |
|
|
|
### Prerequisites |
|
- [Ollama](https://ollama.ai) installed |
|
- 5GB+ free disk space |
|
|
|
### Quick Start |
|
```bash |
|
# Create the model from Modelfile |
|
ollama create rad_cs_myfirstmodel -f Modelfile |
|
|
|
# Run the model |
|
ollama run rad_cs_myfirstmodel |
|
|
|
# Test it out! |
|
>>> Who are you? |
|
I'm rad_cs, your computer science specialist. |
|
|
|
>>> What is binary search? |
|
Binary search is an O(log n) algorithm that efficiently searches sorted arrays... |
|
``` |
|
|
|
## π¬ Example Questions |
|
- **Algorithms**: "Explain quicksort algorithm" |
|
- **Programming**: "Write a Python function to reverse a string" |
|
- **Data Structures**: "How does a hash table work?" |
|
- **Systems**: "What's the difference between TCP and UDP?" |
|
- **Web Dev**: "Create a REST API in Python Flask" |
|
- **AI/ML**: "Explain gradient descent" |
|
|
|
## π οΈ Technical Details |
|
- **Temperature**: 0.7 (balanced creativity/accuracy) |
|
- **Context Window**: 4096 tokens |
|
- **Max Response**: 512 tokens |
|
- **Offline Capable**: Yes, no internet required |
|
- **Privacy**: All processing local, no data sent anywhere |
|
|
|
## π Personality |
|
**Identity Response**: "I'm rad_cs, your computer science specialist." |
|
|
|
**For Non-CS Topics**: Politely redirects to CS subjects: *"I specialize in computer science topics. Please ask about programming, algorithms, systems, or software development."* |
|
|
|
## π§ Model Configuration |
|
The model uses custom system prompts to: |
|
- Focus exclusively on computer science domains |
|
- Provide direct, technical answers |
|
- Include practical code examples |
|
- Maintain educational clarity |
|
- Filter out non-CS topics |
|
|
|
## π My Learning Journey |
|
This represents my first step into AI model creation! Key learnings: |
|
- **No PhD required** - Just vision and open source tools |
|
- **Community support** - Amazing resources available |
|
- **Ollama platform** - Makes local AI accessible |
|
- **Custom specialization** - Turn any model into a domain expert |
|
|
|
## π What's Next? |
|
- Gather community feedback |
|
- Improve based on real usage |
|
- Explore other specializations |
|
- Share knowledge with newcomers |
|
|
|
## π License |
|
MIT License - Feel free to use, modify, and learn from this! |
|
|
|
## π Acknowledgments |
|
- **Open Source Community** - For making this possible |
|
- **Ollama Team** - For the amazing platform |
|
- **Base Model Creators** - For the foundation |
|
- **Early Users** - For feedback and support |
|
|
|
--- |
|
*Built with β€οΈ as my first contribution to the AI community!* |
|
|
|
## π Stats |
|
- **First model**: β
|
|
- **Computer Science focused**: β
|
|
- **Offline privacy**: β
|
|
- **Community ready**: β
|
|
|
|
*What an exciting time to build AI!* π |
|
|