12 Commits

Include renames

Author	SHA1	Message	Date
huangyanpei	d1cc7d96ad	Add Wall-X serving and Turtle2 TCP WebSocket bridge	2026-09-23 21:04:17 +08:00
Starrick Liu	72834e7de5	Update Wall-X to 1.1.0 (#104)	2026-06-15 11:40:00 +08:00
suolyerandyangping	d18fa65fa1	add mot (#83) ... * add mot * update libero example * translate zh to en * fix load model from hf * lint * lint --------- Co-authored-by: yangping <yangping@x2robot.com>	2026-02-03 11:35:25 +08:00
Lufang Chen	d821b0cb26	Fix normalizer (#57) ... * fix normalizer * fix val * update compute stats * delete norm * update readme * minor fix * fix action normalizer * fix * fix * update * update * update * update * update * lint * lint * lint * lint	2025-10-24 17:29:12 +08:00
Lufang Chen	35399d187a	Update train from QwenVL (#50) ... * update from vlm * update * update * update	2025-10-16 10:53:51 +08:00
Lufang Chen	17335bcc3d	Update train test split (#41) ... * update * lint	2025-09-27 12:51:25 +08:00
Starrick Liu	421db17d53	feat: Major optimization and robustness improvements (#31) ... This release introduces significant performance optimizations, memory efficiency improvements, and enhanced system robustness: 🚀 Performance Optimizations: - Add three new fused CUDA kernels (rope_index, rot_pos_emb, get_window_index) for accelerated multimodal preprocessing - Implement FSDP2 support for distributed training with improved memory efficiency - Add Torch.compile integration for additional performance gains - Optimize memory usage: reduce peak allocation from 48GB to 24GB on 8-GPU setup 🔧 System Robustness: - Fix missing token position inputs in prediction pipeline - Add type-robust negation operations in RoPE CUDA kernels (half/bfloat16 support) - Fix dataset root parameter initialization in LeRobot data loader - Enhanced error handling and input validation across fusion operators 📚 Documentation & Usability: - Add comprehensive memory usage benchmarks and hardware recommendations - Update citation format with proper arXiv reference - Improve training configuration documentation with quick start guide - Add detailed API documentation for new fusion operators 🛠️ Technical Details: - Version bump to 1.0.1 - New CUDA kernels: rope_index.cu, rot_pos.cu, window_index.cu - FSDP2 state dict loading with distribute_tensor support - Enhanced multimodal RoPE with 3D position encoding - Window attention optimization for Vision Transformers Breaking Changes: None - all changes are backward compatible	2025-09-17 23:09:20 +08:00
Lufang Chen	e9332a283d	[lint] Update lint (#16) ... * update lint * update readme * update ruff lint	2025-09-11 13:18:33 +08:00
Starrick	f1a9567f61	Fix Readme, fix Timer in training	2025-09-09 15:04:00 +08:00
vincentccc	49af1cf154	fix save check point	2025-09-08 23:21:03 +08:00
vincentchen	ff1feb611f	fix load model	2025-09-08 21:58:59 +08:00
Starrick	24dbdbd24b	Init	2025-09-07 14:59:17 +08:00