twelvedata/financial-world-model
收藏资源简介:
Twelve Data世界模型数据集是一个多模态金融时间序列数据集,基于Twelve Data市场数据构建。数据集包含三种并行视图:bars_*(OHLCV条形图,包含技术指标和宏观背景)、text_*(从bars_*派生的指令调优提示/标签)和trajectories_*(固定长度的滚动窗口状态向量和下一状态对,适合世界模型/序列模型训练)。数据集涵盖相同的符号宇宙和时间分割,适用于时间序列预测、文本生成和强化学习任务。数据集包含51只美国大盘股,覆盖多个行业(科技、金融、医疗、消费、工业/能源/材料、通信)。数据集的刷新频率由增量管道控制,确保数据的及时更新和准确性。
The Twelve Data World Model Dataset is a multi-modal financial time-series dataset built from Twelve Data market data. Each timeframe is published in three parallel views: bars_* (OHLCV bars enriched with causal technical indicators and macro context), text_* (instruction-tuning prompts/labels derived from the bars), and trajectories_* (fixed-length rolling windows of state vectors plus next-state pairs, suitable for world-model / sequence-model training). All views cover the same symbol universe and share the same time-based train / validation / test splits. The dataset includes 51 large-cap US equities spread across sectors (Tech, Financials, Healthcare, Consumer, Industrials/Energy/Materials, Communication). The dataset is refreshed by an incremental pipeline that ensures timely updates and accuracy.




