Code Contribution: 【Lv3】【Operator Development】batch_norm_backward

[StrongSpoon]

Description 任务介绍

Develop backward function for batch_norm operator.
开发batch_norm算子的反向功能。

Requirements 任务要求

Interface 接口
batch_norm_backward(Tensor grad_out, Tensor input, Tensor weight, Tensor? running_mean, Tensor? running_var, Tensor? save_mean, Tensor? save_var, bool update, float eps, bool[3] output_mask, Tensor reserve) -> (Tensor, Tensor, Tensor)
Function reference 功能参考
https://pytorch.org/docs/stable/generated/torch.nn.functional.batch_norm.html
Implementation reference 实现参考
https://github.com/FlagOpen/FlagGems/blob/master/src/flag_gems/ops/groupnorm.py
https://github.com/FlagOpen/FlagGems/blob/master/src/flag_gems/ops/layernorm.py

The operator should support all optional arguments defined in the interface and return the corresponding number of values.
算子应支持接口中定义的所有参数选项，并返回相应数量的值。

DDL 提交时间

Please submit a Pull Request within 3 weeks after accepting the assignment.
请于接取任务后三周内提交PR。
Please provide both accuracy test and performance test code.
请同时提供实现正确性测试与性能测试代码。

[2niuhe]

2niuhe认领