| CARVIEW |
Navigation Menu
-
Notifications
You must be signed in to change notification settings - Fork 5.8k
【Paddle Tensor No.26】fix svdvals #69820
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
|
你的PR提交成功，感谢你对开源项目的贡献! |
|
对了svdvals这个API还麻烦加到中文文档里，可以参考类似的PR：https://github.com/PaddlePaddle/docs/pull/6932/files |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
本地可以打开check_grad，运行一下200x300的单测，python 单测文件.py即可，不需要用CTest，编译也不用开-DWITH_TESTING
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
然后windows的那个编译拆分，代码可以再优化一下
我的本地还在编译着，到时候我测试一下 |
好的好的，我待会修改下 |
好的 |
| start = 0 | ||
| for i in range(n_parts): | ||
| end = min(start + part_size, len(op_list_strs)) | ||
| op_list_parts.append(op_list_strs[start:end]) | ||
| start = end |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
st, ed = i * part_size, min((i+1)*part_size, len(op_list_strs))
| f"op_declare_part_{i + 1}": ",".join(part).replace("\n", "") | ||
| for i, part in enumerate(op_list_parts) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
| f"op_declare_part_{i + 1}": ",".join(part).replace("\n", "") | |
| for i, part in enumerate(op_list_parts) | |
| f"op_declare_part_{i}": ",".join(part).replace("\n", "") | |
| for i, part in enumerate(op_list_parts, 1) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这个文件先保持原样吧，把svdvals的BUG先修了再说，感觉这个文件还是有点问题导致windows inference过不了
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这个文件先保持原样吧，把svdvals的BUG先修了再说，感觉这个文件还是有点问题导致windows inference过不了
是回到只修改了svdvals的版本吗
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这个文件先保持原样吧，把svdvals的BUG先修了再说，感觉这个文件还是有点问题导致windows inference过不了
是回到只修改了svdvals的版本吗
对的，git checkout develop paddle/fluid/pir/dialect/op_generator/op_gen.py 即可
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这个文件先保持原样吧，把svdvals的BUG先修了再说，感觉这个文件还是有点问题导致windows inference过不了
是回到只修改了svdvals的版本吗
对的，
git checkout develop paddle/fluid/pir/dialect/op_generator/op_gen.py即可
这个是两个op_list，需要改回到四个吗
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这个文件先保持原样吧，把svdvals的BUG先修了再说，感觉这个文件还是有点问题导致windows inference过不了
是回到只修改了svdvals的版本吗
对的，
git checkout develop paddle/fluid/pir/dialect/op_generator/op_gen.py即可这个是两个op_list，需要改回到四个吗
现在的develop分支不就是4个吗？我说的是上游的develop
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这个文件先保持原样吧，把svdvals的BUG先修了再说，感觉这个文件还是有点问题导致windows inference过不了
是回到只修改了svdvals的版本吗
对的，
git checkout develop paddle/fluid/pir/dialect/op_generator/op_gen.py即可这个是两个op_list，需要改回到四个吗
现在的develop分支不就是4个吗？我说的是上游的develop
了解，弄错了
| second_part_op_info = op_list_strs[n : 2 * n] | ||
| third_part_op_info = op_list_strs[2 * n : 3 * n] | ||
| fourth_part_op_info = op_list_strs[3 * n :] | ||
| n_parts = 4 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
常量变量名全大写，并且加上必要的注释
| n_parts = 4 | |
| # The number of chunks in the op_declare template | |
| NUM_OP_DECLARE_CHUNK = 4 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
PR Category
User Experience
PR Types
Improvements
Description
fix svdvals