[众智活动任务1]MaxUnpool

CI不通过原因，代码中的一些格式错误

19 KiB

zoulq commented 1 year ago

能够用numpy从原理上实现max_unpool说明对这个接口功能理解深刻，非常厉害！
从adapter实现的角度，建议直接用mindspore实现的接口功能来对标，因为numpy只能运行在cpu侧，而mindspore实现的接口针对GPU/Ascend硬件做了优化，性能上能达到最优。如果两者接口存在参数功能实现不一致，则需要在adapter做适配转换。
mindspore max_unpool1d接口实现介绍参考：https://www.mindspore.cn/docs/zh-CN/master/api_python/ops/mindspore.ops.max_unpool1d.html?highlight=max_unpool1d

zoulq commented 1 year ago

同上，mindspore max_unpool2d接口实现介绍参考：https://www.mindspore.cn/docs/zh-CN/master/api_python/ops/mindspore.ops.max_unpool2d.html?highlight=max_unpool2d

zoulq commented 1 year ago

max_unpool3d也请一并修改。另外CI中错误信息是和代码检查相关(如下)，建议可以直接本地运行 bash ./pylint_check.sh 进行提前检查。

ms_adapter/pytorch/tensor.py:1365:43: I1101: Module 'mindspore._c_expression' has no 'Tensor' member, but source is unavailable. Consider adding this module to extension-pkg-allow-list if you want to perform analysis based on run-time introspection of living objects. (c-extension-no-member)
************* Module ms_adapter.pytorch.nn.functional
ms_adapter/pytorch/nn/functional.py:1464:0: C0325: Unnecessary parens after 'not' keyword (superfluous-parens)
ms_adapter/pytorch/nn/functional.py:1519:0: C0301: Line too long (129/120) (line-too-long)
ms_adapter/pytorch/nn/functional.py:1520:0: C0301: Line too long (125/120) (line-too-long)
ms_adapter/pytorch/nn/functional.py:1545:0: C0301: Line too long (128/120) (line-too-long)

Deleting a branch is permanent. It CANNOT be undone. Continue?

Dear OpenI User

Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.

For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》

zoulq reviewed 1 year ago

ms_adapter/pytorch/nn/functional.py

laich referenced this issue from a commit 1 year ago

[众智活动任务1]MaxUnpool (#196) [众智活动任务1]MaxUnpool Co-authored-by: pengtao <756625088@qq.com> Reviewed-on: https://openi.pcl.ac.cn/OpenI/MSAdapter/pulls/196

laich merged commit 74a44f12fb into master 1 year ago

@@ -1436,0 +1476,4 @@
     indices = cast_to_ms_tensor(indices)
     input_shape = list(input.shape)
     size = _unpool_output_size(input, kernel_size, stride, padding, output_size)

@@ -1436,0 +1493,4 @@
     return out
 def max_unpool2d(input, indices, kernel_size, stride, padding, output_size = None):

@@ -1436,0 +1524,4 @@
     return out.reshape(output_size)
 def max_unpool3d(input, indices, kernel_size, stride, padding, output_size = None):

#196 [众智活动任务1]MaxUnpool