leiyv/marketdata

Polymarket时间序列预测数据集是从Polymarket平台获取的时间序列数据，用于训练WaveNet模型以预测未来市场走势。数据集包含6,780,504个样本，每个样本有1,500个特征维度（100个时间段 × 15个因子）和10个标签维度（预测未来10个时间段）。数据集以Parquet格式存储，总大小约为12.36 GB。数据集分为训练集、验证集和测试集，比例分别为80%、10%和10%。

This is a time-series forecasting dataset extracted from Polymarket, used for training WaveNet models to predict future market trends. The dataset contains 6,780,504 samples, each with 1,500 feature dimensions (100 time periods × 15 factors) and 10 label dimensions (forecasting the next 10 time periods). The dataset is stored in Parquet format, with a total size of approximately 12.36 GB. The dataset is split into training, validation, and test sets, with ratios of 80%, 10%, and 10% respectively.