Skip to content

[API Compatibility] Align paddle.Tensor.is_sparse, paddle.Tensor.type, paddle.Tensor.size api, paddle.nn.PReLU and paddle.distributions.categorical.Categorical -part - #79550

Merged
zhwesky2010 merged 32 commits into
PaddlePaddle:developfrom
Manfredss:three_Tensor_and_PReLU
Aug 13, 2026

Conversation

@Manfredss

@Manfredss Manfredss commented Jul 27, 2026

Copy link
Copy Markdown
Contributor

PR Category

User Experience

PR Types

Improvements

Description

Align the following Paddle APIs with their PyTorch counterparts:

  • Align paddle.Tensor.numel() to return a Python int under enable_compat(level=2).
  • Align paddle.Tensor.is_sparse as a property and return True only for sparse COO tensors.
  • Align paddle.Tensor.type() behavior, including type queries, dtype/class/string conversions, device handling, and same-type identity preservation.
  • Extend paddle.nn.PReLU to accept PyTorch-style device and dtype positional and keyword arguments.
  • Align paddle.distributions.categorical.Categorical

Paddle-internal callers retain native Tensor semantics, and all patched attributes are restored after disabling compat.

Also refine the compat level semantics and make use_compat_guard restore the original compat state. Add lifecycle and real-torch regression tests. This is related to

是否引起精度变化

@paddle-bot

paddle-bot Bot commented Jul 27, 2026

Copy link
Copy Markdown

你的PR提交成功,感谢你对开源项目的贡献!
请关注后续CI自动化测试结果,详情请参考Paddle-CI手册
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

@paddle-bot paddle-bot Bot added the contributor External developers label Jul 27, 2026
@Manfredss
Manfredss marked this pull request as draft July 27, 2026 07:56

@risemeup1111 risemeup1111 left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

序号 位置 优先级 状态
1 python/paddle/utils/decorator_utils.py P2 🟡
2 python/paddle/compat/proxy.py P2
3 python/paddle/compat/__init__.py P1
4 test_project/paddle_temp.py P2
点击展开 Review 规则 PR 评审规则:
  • P0、P1 级别的评审必须提交新的 commit 进行修改;
  • P2、P3 级别的评审可以通过评论进行修改。

状态说明:

已解决 无需修复 待修复
🟡 🚧
Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

Comment thread python/paddle/nn/layer/activation.py Outdated
Comment on lines +589 to +599
is_torch_device = weight_attr is None or (
isinstance(weight_attr, str)
and weight_attr.split(':', 1)[0].lower()
in {'cpu', 'cuda', 'gpu', 'xpu', 'mps', 'meta'}
)
if is_torch_device:
if not isinstance(data_format, str):
device, dtype = weight_attr, data_format
weight_attr, data_format = None, "NCHW"
elif weight_attr is not None and data_format == "NCHW":
device, weight_attr = weight_attr, None

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P1 优先级:P1

问题: 这里把第三个位置参数为 "cpu"/"cuda" 等字符串且 data_format 仍为默认值的调用重解释成 PyTorch 的 device。但 weight_attr: ParamAttrLike 旧 API 明确支持字符串参数名,ParamAttr._to_attr("cpu") 会创建名为 "cpu" 的参数;现在 paddle.nn.PReLU(2, 0.5, "cpu") 会丢掉 weight_attr 并创建匿名参数。

影响: 这是原生 paddle.nn.PReLU 构造函数的全局行为变化,不需要开启 enable_compat(level=2) 就会触发,会破坏依赖参数名的旧模型代码、静态图或 checkpoint/state_dict 兼容性。

处理要求:请针对该评论修复并提交新的 commit。

期望: 保留旧的字符串 weight_attr 语义;PyTorch 的无 dtype 位置参数 device 形态建议放到 paddle.compat.nn.PReLU/level=2 alias 中处理,或至少只在不会与旧 weight_attr 字符串冲突的形态下重解释。例如可以按下面的形态收窄 native 构造函数里的自动重解释:

# 伪代码:仅处理第四个位置参数明确是 dtype/device pair 的无冲突形态
if not isinstance(data_format, str):
    device, dtype = weight_attr, data_format
    weight_attr, data_format = None, "NCHW"

# PReLU(..., "cpu") 这种无 dtype 的 PyTorch 形态不要在 native paddle.nn.PReLU
# 中覆盖字符串 weight_attr;如需支持,请放到 compat-only wrapper 中。

@codecov-commenter

codecov-commenter commented Jul 27, 2026

Copy link
Copy Markdown

Codecov Report

❌ Patch coverage is 91.77215% with 13 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (develop@76d6cef). Learn more about missing BASE report.

Files with missing lines Patch % Lines
python/paddle/compat/proxy.py 68.96% 9 Missing ⚠️
python/paddle/compat/__init__.py 95.45% 3 Missing ⚠️
python/paddle/utils/decorator_utils.py 94.44% 1 Missing ⚠️
Additional details and impacted files
@@            Coverage Diff             @@
##             develop   #79550   +/-   ##
==========================================
  Coverage           ?   91.77%           
==========================================
  Files              ?        6           
  Lines              ?      158           
  Branches           ?        0           
==========================================
  Hits               ?      145           
  Misses             ?       13           
  Partials           ?        0           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

@Manfredss

Copy link
Copy Markdown
Contributor Author

/re-run all-failed

@Manfredss
Manfredss requested a review from risemeup1111 July 27, 2026 13:07
@Manfredss

Copy link
Copy Markdown
Contributor Author

/re-run all-failed

@Manfredss
Manfredss marked this pull request as ready for review July 28, 2026 02:11

@risemeup1111 risemeup1111 left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

已复查当前 head,先前指出的 PReLU(2, 0.5, "cpu") 后向兼容问题在代码上已经修复。本次新增一个非阻塞的行级建议,细节见 inline comment;另外 CI 里 Static-Check / Slice 仍有红灯,合入前还需要继续跟进。

Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

Comment thread python/paddle/compat/__init__.py Outdated
return f"{prefix}.{tensor_type}"

device = None
if isinstance(dtype, type):

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 优先级:P2

问题: 这里用 isinstance(dtype, type) 捕获所有 Python type,然后只按 FloatTensor/DoubleTensor 这类 tensor factory 名称识别。DTypeLike 里包含 np.float32np.float64np.int64 等 numpy scalar type;这些对象也是 type,会在这里因为 dtype.__name__(例如 "float64")不在 _TENSOR_TYPE_DTYPES 而直接 ValueError,无法走后面的 input.to(dtype=dtype)

影响: enable_compat(level=2) 后,t.type(np.float64) 这类 Paddle 其他 dtype 参数常见写法会被误判为非法 tensor class。与当前签名里的 DTypeLike | str | type 不一致,也缺少测试覆盖。

处理要求:请针对该评论进行回复(同意并已修改请回复 Done,不同意请说明理由)。

期望: 只把已知 tensor factory class 当作 PyTorch-style class conversion,其他 DTypeLike 继续交给 Tensor.to 校验/转换,并补一个 np.float64 之类的回归测试。可以先把该行收窄为:

Suggested change
if isinstance(dtype, type):
if isinstance(dtype, type) and dtype.__name__ in _TENSOR_TYPE_DTYPES:

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 优先级:P2

补充复查:新提交的 test_tensor_type_edge_cases 覆盖了 t.type(paddle.float64, **{"async": True}),但这个参数不是 Python type,不会进入当前 if isinstance(dtype, type): 分支;原问题里的 np.float64 / np.int64 等 numpy scalar type 仍会进入该分支并因为 "float64" 不在 _TENSOR_TYPE_DTYPES 里抛 ValueError

处理要求:请针对该评论进行回复(同意并已修改请回复 Done,不同意请说明理由)。

建议仍按上面 suggestion 收窄 tensor factory class 的判断,并把回归测试补到 numpy dtype 路径,例如:

import numpy as np

self.assertEqual(t.type(np.float64).dtype, paddle.float64)

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P1 优先级:P1

补充复查:这个提交把原来的判断改成了 dtype.__name__ in _TENSOR_TYPE_NAMES,但 _TENSOR_TYPE_NAMES 的 key 是 "float32"/"float64" 这类 dtype 名;paddle.DoubleTensor 来自 dtype_tensor_factory('float64', 'DoubleTensor')__name__"DoubleTensor"。因此 t.type(paddle.DoubleTensor) 不会进入这里的 tensor factory class 分支,会继续把 class 对象传给 input.to(dtype=...),这不是 DTypeLike 支持的输入,新增的 self.assertEqual(t.type(paddle.DoubleTensor).dtype, paddle.float64) 也会失败。

处理要求:请针对该评论修复并提交新的 commit。

请把 class 分支的 membership 查 _TENSOR_TYPE_DTYPES(反查表)而不是 _TENSOR_TYPE_NAMES,并保留这次补上的 np.float64 回归测试。修复形态可以是:

if isinstance(dtype, type) and dtype.__name__ in _TENSOR_TYPE_DTYPES:
    tensor_type = dtype.__name__
    dtype = _TENSOR_TYPE_DTYPES[tensor_type]
    device = "cpu"

@risemeup1111 risemeup1111 left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

已复查最新提交。先前的 PReLU(2, 0.5, "cpu") 后向兼容问题在当前代码里仍是修复状态。

不过 _tensor_type 的 follow-up 修复把 tensor factory class 的查表对象写反了,当前仍会影响 t.type(paddle.DoubleTensor) 这类兼容路径;我已在既有行级线程里补充了具体证据和修复形态。这个问题需要再提交一个修复 commit 后才能通过。当前还有部分 CI 在运行中,请合入前继续确认结果。

Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

@risemeup1111 risemeup1111 left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

已复查最新提交,先前关于 PReLU 后向兼容和 Tensor.type dtype/class 处理的意见在当前代码中都已修复,未发现新的需要阻塞合入的问题。当前仍有部分 CI 在运行中,合入前请继续确认最终结果。

Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

@Manfredss

Copy link
Copy Markdown
Contributor Author

/re-run all-failed

1 similar comment
@Manfredss

Copy link
Copy Markdown
Contributor Author

/re-run all-failed

@risemeup1111 risemeup1111 left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

已复查最新提交。此前关于 PReLUTensor.type 的问题在当前代码中仍是修复状态。

这次新增的 Categorical 参数校验还有一个需要修正的兼容性问题,具体见行级评论。当前仍有部分 CI 在运行中,合入前请继续确认最终结果。

Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

Comment thread python/paddle/compat/distributions/categorical.py
@Manfredss

Copy link
Copy Markdown
Contributor Author

/re-run all-failed

@risemeup1111 risemeup1111 left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

已复查你在 Categorical 线程里的说明,确认当前实现与 PyTorch 的 probs 归一化/校验顺序一致;此前这条阻塞意见已撤回。先前关于 PReLUTensor.type 的问题在当前 head 中也仍是修复状态,未发现新的需要阻塞合入的问题。

当前仍有部分 CI 未完成/有红灯,请合入前继续确认最终结果。

Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

@Manfredss
Manfredss requested a review from risemeup1111 July 29, 2026 02:37
@Manfredss Manfredss changed the title [API Compatibility] Align paddle.Tensor.is_sparse, paddle.Tensor.type paddle.Tensor.size api and paddle.nn.PReLU -part [API Compatibility] Align paddle.Tensor.is_sparse, paddle.Tensor.type, paddle.Tensor.size api, paddle.nn.PReLU and paddle.distributions.categorical.Categorical -part Jul 29, 2026

@risemeup1111 risemeup1111 left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

已复查标题更新,当前标题已覆盖这次新增的 Categorical 兼容范围;代码 head 未变化,先前评审结论不变,未发现新的需要阻塞合入的问题。

当前仍有部分 CI 未完成/有红灯,请合入前继续确认最终结果。

Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

@Manfredss

Copy link
Copy Markdown
Contributor Author

/re-run all-failed

@Manfredss

Manfredss commented Jul 29, 2026

Copy link
Copy Markdown
Contributor Author

@zhwesky20 review this PR, Thx

@zhwesky2010

Copy link
Copy Markdown
Contributor

@Manfredss 机器人这个CI能通过吗

@Manfredss

Manfredss commented Aug 10, 2026

Copy link
Copy Markdown
Contributor Author

@Manfredss 机器人这个CI能通过吗

@zhwesky2010

#79550 (comment)

#79550 (comment)

我觉得是不需要改这个地方的,从最原始的版本开始(添加 level 前)就是这样实现的,且对运行没有影响

@Manfredss
Manfredss requested a review from risemeup1111 August 10, 2026 08:45
@zhwesky2010

zhwesky2010 commented Aug 10, 2026

Copy link
Copy Markdown
Contributor

@Manfredss 机器人这个CI能通过吗

@zhwesky2010

#79550 (comment)

#79550 (comment)

我觉得是不需要改这个地方的,从最原始的版本开始(添加 level 前)就是这样实现的,且对运行没有影响

这个你需要和机器人沟通,直到机器人approve。不然CI是强制不让过的

@Manfredss

Copy link
Copy Markdown
Contributor Author

/re-run all-failed

@Manfredss

Copy link
Copy Markdown
Contributor Author

当前 head 仍未修复前次指出的两处 P1:paddle.nn.PReLU 仍会把字符串位置参数当作 device 重解释,enable_compat() / use_compat_guard(enable=False) 的 finder 恢复也还没有改成幂等插入或完整恢复。因此我仍然请求修改。

Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

@risemeup1111 还有什么地方没有通过?

@risemeup1111

Copy link
Copy Markdown
Contributor

目前只剩 1 项未通过:enable_compat() / use_compat_guard(enable=False) 的 finder 生命周期。PReLU 项已结合实际使用情况接受,不再阻塞。

当前 enable_compat() 重复调用会把同一 TORCH_PROXY_FINDER 插入两次,而 use_compat_guard(enable=False) 只调用一次 disable_compat();因此 guard 内 finder 数量会从 2 变为 1,compat 并未真正禁用。请将插入改为幂等并同步测试为 1 -> 0 -> 1 -> 0,或者在 guard 中完整清空并按原数量恢复。修复这一项后我会复核当前 head。

@Manfredss

Manfredss commented Aug 10, 2026

Copy link
Copy Markdown
Contributor Author

目前只剩 1 项未通过:enable_compat() / use_compat_guard(enable=False) 的 finder 生命周期。PReLU 项已结合实际使用情况接受,不再阻塞。

当前 enable_compat() 重复调用会把同一 TORCH_PROXY_FINDER 插入两次,而 use_compat_guard(enable=False) 只调用一次 disable_compat();因此 guard 内 finder 数量会从 2 变为 1,compat 并未真正禁用。请将插入改为幂等并同步测试为 1 -> 0 -> 1 -> 0,或者在 guard 中完整清空并按原数量恢复。修复这一项后我会复核当前 head。

@risemeup1111 两位审核已经同意此次不修改这个地方,已经显式 approve;且当前并不影响功能与使用,从最原始的版本开始(添加 level 前)就是这样实现的

@risemeup1111

Copy link
Copy Markdown
Contributor

你说得对。我重新对比了 base head 与当前实现:重复 enable_compat() 插入多个 finder、use_compat_guard(enable=False) 只移除一次的组合在本 PR 前已经存在,本 PR 没有引入或扩大该路径;仓库和公开代码检索也只发现文档/测试,没有实际业务调用这一组合。结合两位 API reviewer 已接受现有语义,我将该项调整为 P2 并接受当前方案。现在没有剩余阻塞项,我会更新 Board 并 approve 当前 head。

@Manfredss

Copy link
Copy Markdown
Contributor Author

@zhwesky2010

@zhwesky2010 zhwesky2010 left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

fleet unittest这个不确定是不是PR的问题,其他PR暂时没遇到超时问题。

拉下develop重新push下看看,不排除是PR问题导致超时,fleet中很多地方用了enable_compat

Comment thread python/paddle/compat/__init__.py Outdated
raise TypeError(f"type() got an unexpected keyword argument {key!r}")

if dtype is None:
return str(input.dtype)

@zhwesky2010 zhwesky2010 Aug 11, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

你这个地方需要映射为torch一致的paddle.FloatTensor这种格式

@Manfredss Manfredss Aug 11, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这个是 paddle.FloatTensor 还是 torch.FloatTensor? paconvert 那边比较的话是拿 paddle 的返回值和 torch 的返回值比对的,torch 返回 'torch.FloatTensor', 所以我觉得这里也用 torch 吧,但确实 paddle 里调用返回 torch 很怪

>>> import torch
>>> a = torch.tensor([2.], dtype=float32)
>>> a.type()
'torch.FloatTensor'

然后像 uint16 这些直接返回字符串的,返回的也是 paddle.uint16 这样的,需要拼成 torch 前缀吗

@Manfredss Manfredss Aug 11, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@zhwesky2010

还有这个问题 #79641 (comment)

我开了一个测试 PR 修这个问题,如果这样修改没问题的话我再推这个 PR 上

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

我的想法是另开一个 PR 修这个问题,要涉及到修改 creation.py,并且还有是 torch 还是 paddle 前缀的问题

@Manfredss Manfredss Aug 12, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

我的想法是另开一个 PR 修这个问题,要涉及到修改 creation.py,并且还有是 torch 还是 paddle 前缀的问题

我认为 #79641 这样的修改是对的

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

当前这项仍需在本 PR 修复。compat 模式下 t.type() 仍返回 paddle.float32,而本 PR 描述明确声明对齐 PyTorch 的 type query;CPU float32 应返回 torch.FloatTensor,GPU 与 sparse COO 还需编码 place/layout。现有测试反而将 paddle.float32 固化为预期值。请将 #79641 中已确认的类型名生成逻辑及 CPU/GPU/稀疏回归测试合入本 PR 后再合入;单独后续修复会使本 PR 先发布一个与声明不一致的公共行为。

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

当前提交已修复 CPU、CUDA 和 sparse COO 的查询,但这项仍是部分修复:_tensor_type_name() 只判断 is_gpu_place(),XPU/custom place 都会返回 paddle.FloatTensor,再调用 t.type(t.type()) 时该字符串会按 CPU 类型解析并迁移设备。请像 #79641 的后续实现一样为 XPU 和 custom device 保留设备段,并补对应 fake-place/round-trip 测试;完成后这项即可关闭。

Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

已确认当前提交为 XPU 和已注册 custom device 保留了设备段,字符串解析与 no-op 判断也使用相同的 place 映射;新增测试覆盖了 CPU/GPU/XPU/custom 输出、有效转换和非法多设备段。此前的设备 round-trip 问题已解决。

Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

@Manfredss
Manfredss dismissed stale reviews from SigureMo and zhwesky2010 via 83282be August 13, 2026 01:52
Comment thread test_project/paddle_temp.py Outdated

import paddle

paddle.enable_compat(level=2)

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 这两个 test_project/*_temp.py 是未被任何测试入口引用的手工对照脚本,也没有断言,CI 不会执行其中的 float8 检查;对应行为已经在 test/compat/test_compat_namespace_aliased.py 中覆盖。请删除这两个临时文件,避免把不可执行的调试产物作为回归测试提交。

Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

已删除

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

已确认当前提交删除了这两个临时脚本,此项已解决。

Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

Comment thread python/paddle/compat/__init__.py Outdated
raise TypeError(f"type() got an unexpected keyword argument {key!r}")

if dtype is None:
return str(input.dtype)

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

当前提交已修复 CPU、CUDA 和 sparse COO 的查询,但这项仍是部分修复:_tensor_type_name() 只判断 is_gpu_place(),XPU/custom place 都会返回 paddle.FloatTensor,再调用 t.type(t.type()) 时该字符串会按 CPU 类型解析并迁移设备。请像 #79641 的后续实现一样为 XPU 和 custom device 保留设备段,并补对应 fake-place/round-trip 测试;完成后这项即可关闭。

Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

if has_paddle_aliases:
_apply_paddle_namespace_aliases()
else:
disable_compat()

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

当前提交已同步 develop 的 compat 生命周期实现:enable_compat() 对 finder 采用幂等插入,guard 也按原 level 恢复。因此此前的重复 finder/禁用不完整问题已解决,此线程可以关闭。

Powered by Nyanpasu with gpt-5.5 xhigh, please check the suggestions carefully.

@Manfredss

Copy link
Copy Markdown
Contributor Author

/re-run all-failed

@Manfredss
Manfredss requested a review from zhwesky2010 August 13, 2026 07:14
@Manfredss

Copy link
Copy Markdown
Contributor Author

/re-run all-failed

@Manfredss
Manfredss requested a review from SigureMo August 13, 2026 07:53
@zhwesky2010
zhwesky2010 merged commit 4ab707b into PaddlePaddle:develop Aug 13, 2026
226 of 233 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants