henryzhang2024/qs-deepseek-platinum-v3

---
license: mit
task_categories:
- text-generation
- text2text-generation
language:
- en
tags:
- finance
- trading
- options
- quantitative
size_categories:
- 10K<n<100K
---

# QS DeepSeek Platinum v3

High-quality training dataset for DeepSeek trading model.

## Dataset Details

- **Total samples:** 17,212
- **Format:** OpenAI messages format (system/user/assistant)
- **Sources:**
- qs-deepseek-platinum-15k (HF)
- Synthetic negative examples
- Live signals from Supabase

## Data Quality

| Metric | Value |
|--------|-------|
| Positive PnL | 74.4% |
| Negative PnL | 25.5% |
| Avg message length | 2,220 chars |
| Max message length | 8,922 chars |

## Label Distribution

- Neutral/Hold: 35.9%
- Strong Buy: 20.1%
- Strong Sell: 16.2%
- Buy: 13.0%
- Sell: 10.4%

## Improvements from v2

1. **More negative examples** (25.5% vs 2.3%)
2. **Larger dataset** (17k vs 13k)
3. **Length filtered** (max 9k chars)
4. **Merged with live data**

## Usage

```python
from datasets import load_dataset

dataset = load_dataset("henryzhang2024/qs-deepseek-platinum-v3")
```

## License

MIT