From 6c182ce95968c71571e0f25bfddba31a3b4bcd8b Mon Sep 17 00:00:00 2001 From: cutetocute Date: Wed, 11 Feb 2026 16:42:57 +0800 Subject: [PATCH] [Docs] Fix invalid api_label references --- docs/api/paddle/device/Overview_cn.rst | 4 ++-- docs/api/paddle/gather_cn.rst | 2 +- docs/api/paddle/linalg/Overview_cn.rst | 2 +- docs/api/paddle/linalg/eigvals_cn.rst | 2 +- docs/api/paddle/linalg/matrix_norm_cn.rst | 2 +- docs/api/paddle/linalg/vector_norm_cn.rst | 2 +- docs/api_guides/low_level/layers/sparse_update.rst | 4 ++-- docs/api_guides/low_level/layers/sparse_update_en.rst | 4 ++-- 8 files changed, 11 insertions(+), 11 deletions(-) diff --git a/docs/api/paddle/device/Overview_cn.rst b/docs/api/paddle/device/Overview_cn.rst index 3a76b17d7de..6949846188f 100644 --- a/docs/api/paddle/device/Overview_cn.rst +++ b/docs/api/paddle/device/Overview_cn.rst @@ -46,9 +46,9 @@ paddle.device 目录下包含 cuda 目录和 xpu 目录， cuda 目录中存放 " :ref:`is_available ` ", "检查设备是否可用" " :ref:`get_rng_state ` ", "获取随机数生成器状态" " :ref:`set_rng_state ` ", "设置随机数生成器状态" - " :ref:`device <_cn_api_paddle_device_device>` ", "临时使用设备" + " :ref:`device ` ", "临时使用设备" " :ref:`get_device_name ` ", "返回指定设备的名称" - " :ref:`manual_seed <_cn_api_paddle_device_manual_seed>` ", "设置当前设备的随机数种子" + " :ref:`manual_seed ` ", "设置当前设备的随机数种子" .. _cn_device_compile: 编译环境检测 diff --git a/docs/api/paddle/gather_cn.rst b/docs/api/paddle/gather_cn.rst index 39243f7974c..618f8b054be 100644 --- a/docs/api/paddle/gather_cn.rst +++ b/docs/api/paddle/gather_cn.rst @@ -59,7 +59,7 @@ COPY-FROM: paddle.gather .. py:function:: paddle.gather(input, dim, index, out=None) -PyTorch 兼容的 ``gather`` 操作：根据索引 index 获取输入 ``input`` 的指定 ``dim`` 维度的条目，并将它们拼接在一起。行为与 ``cn_api_paddle_take_along_axis`` 在 ``broadcast=False`` 情况下一致。 +PyTorch 兼容的 ``gather`` 操作：根据索引 index 获取输入 ``input`` 的指定 ``dim`` 维度的条目，并将它们拼接在一起。行为与 :ref:`cn_api_paddle_take_along_axis` 在 ``broadcast=False`` 情况下一致。 接口对比可见 `【torch 参数更多】torch.gather`_ 。 diff --git a/docs/api/paddle/linalg/Overview_cn.rst b/docs/api/paddle/linalg/Overview_cn.rst index 7a241c9b815..97d4b557dae 100644 --- a/docs/api/paddle/linalg/Overview_cn.rst +++ b/docs/api/paddle/linalg/Overview_cn.rst @@ -64,7 +64,7 @@ paddle.linalg 目录下包含飞桨框架支持的线性代数相关 API。具 " :ref:`paddle.linalg.cholesky ` ", "计算一个实数对称正定矩阵的 Cholesky 分解" " :ref:`paddle.linalg.cholesky_inverse ` ", "使用 Cholesky 因子 `U` 计算对称正定矩阵的逆矩阵" " :ref:`paddle.linalg.svd ` ", "计算矩阵的奇异值分解" - " :ref:`paddle.linalg.svdvals <_cn_api_paddle_linalg_svdvals>` ", "计算矩阵的奇异值" + " :ref:`paddle.linalg.svdvals ` ", "计算矩阵的奇异值" " :ref:`paddle.linalg.svd_lowrank ` ", "对低秩矩阵进行奇异值分解" " :ref:`paddle.linalg.pca_lowrank ` ", "对矩阵进行线性主成分分析" " :ref:`paddle.linalg.qr ` ", "计算矩阵的正交三角分解（也称 QR 分解）" diff --git a/docs/api/paddle/linalg/eigvals_cn.rst b/docs/api/paddle/linalg/eigvals_cn.rst index 627db6093e8..6f8b99f43dc 100644 --- a/docs/api/paddle/linalg/eigvals_cn.rst +++ b/docs/api/paddle/linalg/eigvals_cn.rst @@ -8,7 +8,7 @@ eigvals .. note:: - 该 API 的反向实现尚未完成，若你的代码需要对其进行反向传播，请使用 ref:`cn_api_paddle_linalg_eig`。 + 该 API 的反向实现尚未完成，若你的代码需要对其进行反向传播，请使用 :ref:`cn_api_paddle_linalg_eig`。 参数 diff --git a/docs/api/paddle/linalg/matrix_norm_cn.rst b/docs/api/paddle/linalg/matrix_norm_cn.rst index a5197bdae80..ad9fda839b6 100644 --- a/docs/api/paddle/linalg/matrix_norm_cn.rst +++ b/docs/api/paddle/linalg/matrix_norm_cn.rst @@ -8,7 +8,7 @@ matrix_norm -将计算给定 Tensor 的矩阵范数。具体用法请参见 :ref:`norm <_cn_api_paddle_linalg_norm>`。 +将计算给定 Tensor 的矩阵范数。具体用法请参见 :ref:`norm `。 参数 diff --git a/docs/api/paddle/linalg/vector_norm_cn.rst b/docs/api/paddle/linalg/vector_norm_cn.rst index bd67e735c0d..7e8365664cc 100644 --- a/docs/api/paddle/linalg/vector_norm_cn.rst +++ b/docs/api/paddle/linalg/vector_norm_cn.rst @@ -8,7 +8,7 @@ vector_norm -将计算给定 Tensor 的向量范数。具体用法请参见 :ref:`norm <_cn_api_paddle_linalg_norm>`。 +将计算给定 Tensor 的向量范数。具体用法请参见 :ref:`norm `。 参数 diff --git a/docs/api_guides/low_level/layers/sparse_update.rst b/docs/api_guides/low_level/layers/sparse_update.rst index 0765952993f..f4f8a6f48fc 100644 --- a/docs/api_guides/low_level/layers/sparse_update.rst +++ b/docs/api_guides/low_level/layers/sparse_update.rst @@ -4,7 +4,7 @@ 稀疏更新 ##### -`paddle.nn.functional.embedding ` 层在单机训练和分布式训练时，均可以支持“稀疏更新”，即梯度以 sparse tensor 结构存储，只保存梯度不为 0 的行。 +:ref:`paddle.nn.functional.embedding ` 层在单机训练和分布式训练时，均可以支持“稀疏更新”，即梯度以 sparse tensor 结构存储，只保存梯度不为 0 的行。 在分布式训练中，对于较大的 embedding 层，开启稀疏更新有助于减少通信数据量，提升训练速度。 在 paddle 内部，我们用 lookup_table 来实现 embedding。下边这张图说明了 embedding 在正向和反向计算的过程： @@ -14,4 +14,4 @@ .. image:: ../../../design/dist_train/src/lookup_table_training.png :scale: 50 % -API 详细使用方法参考 `paddle.nn.functional.embedding ` +API 详细使用方法参考 :ref:`paddle.nn.functional.embedding ` diff --git a/docs/api_guides/low_level/layers/sparse_update_en.rst b/docs/api_guides/low_level/layers/sparse_update_en.rst index 36fa345ef1a..7b38a759295 100755 --- a/docs/api_guides/low_level/layers/sparse_update_en.rst +++ b/docs/api_guides/low_level/layers/sparse_update_en.rst @@ -4,7 +4,7 @@ Sparse update ############### -`paddle.nn.functional.embedding ` layer supports "sparse updates" in both single-node and distributed training, which means gradients are stored in a sparse tensor structure where only rows with non-zero gradients are saved. +:ref:`paddle.nn.functional.embedding ` layer supports "sparse updates" in both single-node and distributed training, which means gradients are stored in a sparse tensor structure where only rows with non-zero gradients are saved. In distributed training, for larger embedding layers, sparse updates reduce the amount of communication data and speed up training. In paddle, we use lookup_table to implement embedding. The figure below illustrates the process of embedding in the forward and backward calculations: @@ -17,4 +17,4 @@ As shown in the figure: two rows in a Tensor are not 0. In the process of forwar Example -------------------------- -API reference `paddle.nn.functional.embedding ` . +API reference :ref:`paddle.nn.functional.embedding ` .