Repository Issue Activity (beta)

#850 Dataset_M4 数据读取有 bug

#855 时间有隔断怎么处理，比如每天有8个小时是没有数据的

#854 排行榜标准问题

#847 长时间预测训练效果不及预期

#852 [Enhancement] Add Temporal Patch Shuffle (TPS) augmentation for forecasting and classification

#848 📋 Documentation Enhancement Suggestion

#846 Timemixer的通道独立问题

#843 data_provider的shuffle_flag

#839 TimeMixer++ implementation is missing

#838 PatchTST/TSMixer scripts/configs differ from official paper implementation.

#837 Proposal: Adding T1 (ICLR 2026) to the Library

#835 [Bug] Duplicate `datasets` dependency with conflicting versions in requirements.txt

#834 Clarification on data split offset (- seq_len) in data loader

#833 weather重复时间戳

#832 关于zero-shot与finetune

#779 [多进程优化] 在脚本中尝试加入对numpy多线程限制

#750 layer sizes to compare iTransformer vs Vanilla Transformers

#776 About Adjustment

#761 使用混合精度amp训练 报错 xf = torch.fft.rfft(x, dim=1) RuntimeError: cuFFT only supports dimensions whose sizes are powers of two when computing in half precision, but got a signal size of[192]

#795 SMD数据集的默认步长不统一

#799 维度不一致：Dimensions do not match

#826 TimeMixer 的DFT分解

#825 The load_dataset function is missing

#808 版本依赖requirement.txt还是不对，这个版本下的torch都没有

#815 用了--use_gpu False還是跑去使用gpu