模型
120M参数的时间序列基础模型，支持零样本预测与LoRA微调。
已跟踪 2 条高相关材料
最近变化
2026-06-04 · 使用LoRA冻结120M参数主模型，仅训练低秩适配器以高效适配私有数据。
为什么值得关注
Chronos-2 被反复提及时，通常意味着它正在影响产品路线、开发者工作流或 AI 产业判断。这个页面把分散材料合并成一个可持续更新的观察入口。
Five Questions About Chronos-2, the Time Series Foundation Model
Towards Data Science · 8.5 分
Chronos-2 is a time series foundation model that can be used for various forecasting tasks without retraining, providing predictive quantil...
Five Ways to Fine-Tune Chronos-2, the Time Series Foundation Model
Towards Data Science · 8.2 分
Chronos-2时间序列基础模型可通过LoRA微调解决零样本预测偏差，文章详解了单建筑适配、组合池化、协变量注入等五种实战场景及数据划分规范。
已收录 2 条与 Chronos-2 相关的内容，按评分排序。
Chronos-2 is a time series foundation model that can be used for various forecasting tasks without retraining, providing predictive quantiles and working with limited data.
入选理由：Chronos-2 is a pretrained time series model.
Chronos-2 TSFM can be fine-tuned via LoRA to address zero-shot gaps, detailing five scenarios including single-building adaptation, portfolio pooling, and covariate injection with strict data splitting.
入选理由：使用LoRA冻结120M参数主模型，仅训练低秩适配器以高效适配私有数据。
