co63oc

Results 142 comments of co63oc

[0-size Tensor No.91、308] Add 0-size Tensor support for inner

/re-run all-failed

[0-size Tensor Job2 No.100] Add 0-size Tensor support for topk

/re-run all-failed

[0-size Tensor Job2 No.100] Add 0-size Tensor support for topk

/re-run all-failed

[0-size Tensor Job2 No.100] Add 0-size Tensor support for topk

@luotao1 CI已完成

[0-size Tensor No.180、182-183] Add 0-size Tensor support for hinge_embedding_loss

@DanielSun11 CI已完成请review

[0-size Tensor No.180、182-183] Add 0-size Tensor support for hinge_embedding_loss

/re-run all-failed

[0-size Tensor No.180、182-183] Add 0-size Tensor support for hinge_embedding_loss

> 目前paddleAPITest支持大部分api和torch的对比，请拉取最新的paddleAPITest并使用-- accuracy=True来和torch进行对比。 除了不支持对比的api，paddle_only = True的回测结果已经没有太大的参考价值。特别是需要填充默认值的情况下，需要torch对比一下，以防止后续回测继续报错。现有的测试中不该再出现paddle error，paddle_only pass的case也需要和torch对比一下 已复测更新了描述

[0-size Tensor No.180、182-183] Add 0-size Tensor support for hinge_embedding_loss

> paddle.nn.functional.kl_div 中CPU[not compare grad]是因为反向返回了一个None，这是存在问题的，因为GPU中有反向cpu中应该也有反向，不应为None。 有两个返回值，一个为None，测试需要在PaddleAPITest增加配置 https://github.com/PFCCLab/PaddleAPITest/pull/268

Rename ctx to dev_ctx in paddle/phi/kernels/fusion/xpu/ directory

/re-run all-failed

Rename ctx to dev_ctx in paddle/phi/kernels/fusion/xpu/ directory

/re-run all-failed