kreasof-ai/KAC-QuantFin-1M

Kreasof AI Capital (KAC) QuantFin-1M数据集使用基于GARCH模型的合成价格生成方法和公司专有的交易算法。原始的合成价格路径包含28800个步骤，每个步骤代表1分钟，但为了节省存储空间，被降采样为1028个步骤，每个步骤代表28分钟。数据集包含scenario_id、final_pnl_ratio、max_drawdown、total_trades、synthetic_ohlcv_close等字段，并分为训练集，共有大约100万个示例。The Kreasof AI Capital (KAC) QuantFin-1M dataset uses GARCH-based synthetic price generation combined with the companys proprietary trading algorithm. The original synthetic price path consists of 28800 steps at 1-minute intervals, but it has been downscaled to 1028 steps at 28-minute intervals to save storage. The dataset includes fields such as scenario_id, final_pnl_ratio, max_drawdown, total_trades, synthetic_ohlcv_close, and is split into a training set with approximately 1 million examples.