PanZezhong1725 7f6aec6c17 针对bert和gpt2模型分布式推理的优化 (#221) ... * fix(dist): 改善分布式脚本，只打印绝对误差 * feat(dist): 增加可导出onnx的pytorch运行脚本 * feat(front): 增加对Y值为-inf的where算子的图优化 * feat(kernel): 对b为常数的pow和div算子进行特判优化 * fix(front): 消除前端对global output形状信息的依赖，分布式脚本删除不必要的shape infer * feat(kernel): 针对matmul中bias为行向量时的expand操作的特化优化 * fix(kernel): 删除div pow const中不必要的同步 * Update expand.cu * fix: fix comments --------- Co-authored-by: Haojie Wang <haojie0429@gmail.com> Co-authored-by: Derui Yang <ydrml@hotmail.com>		2024-04-01 14:04:28 +08:00
..
NNmodel@51d3105277	XCCL support (#171)	2024-02-29 11:48:35 +08:00
distributed	针对bert和gpt2模型分布式推理的优化 (#221)	2024-04-01 14:04:28 +08:00
python	[feature] support kvcache with static graph (#209)	2024-01-25 14:20:43 +08:00