Fix tenosr tensor (PaddlePaddle#61129)

* Fix

* Fix

Author: co63oc

paddle/fluid/operators/detection/locality_aware_nms_op.cc

@@ -495,7 +495,7 @@ threshold NMS based on parameters of nms_threshold and nms_eta.
 Aftern NMS step, at most keep_top_k number of total bboxes are to be kept
 per image if keep_top_k is larger than -1.
 This operator support multi-class and batched inputs. It applying NMS
-independently for each class. The outputs is a 2-D LoDTenosr, for each
+independently for each class. The outputs is a 2-D LoDTensor, for each
 image, the offsets in first dimension of phi::DenseTensor are called LoD, the number
 of offset is N + 1, where N is the batch size. If LoD[i + 1] - LoD[i] == 0,
 means there is no detected bbox for this image.

paddle/fluid/operators/detection/multiclass_nms_op.cc

@@ -562,7 +562,7 @@ threshold NMS based on parameters of nms_threshold and nms_eta.
 Aftern NMS step, at most keep_top_k number of total bboxes are to be kept
 per image if keep_top_k is larger than -1.
 This operator support multi-class and batched inputs. It applying NMS
-independently for each class. The outputs is a 2-D LoDTenosr, for each
+independently for each class. The outputs is a 2-D LoDTensor, for each
 image, the offsets in first dimension of phi::DenseTensor are called LoD, the number
 of offset is N + 1, where N is the batch size. If LoD[i + 1] - LoD[i] == 0,
 means there is no detected bbox for this image.

paddle/fluid/operators/detection/retinanet_detection_output_op.cc

@@ -648,7 +648,7 @@ threshold NMS based on parameters of nms_threshold and nms_eta.
 After NMS step, at most keep_top_k number of total bounding boxes are to be kept
 per image if keep_top_k is larger than -1.
 This operator support multi-class and batched inputs. It applying NMS
-independently for each class. The outputs is a 2-D LoDTenosr, for each
+independently for each class. The outputs is a 2-D LoDTensor, for each
 image, the offsets in first dimension of phi::DenseTensor are called LoD, the number
 of offset is N + 1, where N is the batch size. If LoD[i + 1] - LoD[i] == 0,
 means there is no detected bounding box for this image. If there is no detected boxes

paddle/phi/infermeta/binary.cc

@@ -1279,7 +1279,7 @@ void DotInferMeta(const MetaTensor& x, const MetaTensor& y, MetaTensor* out) {
       x_rank == static_cast<int>(y_dims.size()),
       phi::errors::PreconditionNotMet(
           "ShapeError: The shape of input tensor Y: %s should match with "
-          "input tenosr X: %s",
+          "input tensor X: %s",
           y_dims.to_str(),
           x_dims.to_str()));
   bool shape_match = true;

paddle/phi/infermeta/spmd_rules/utils.cc

@@ -27,7 +27,7 @@ namespace distributed {
 
 using phi::distributed::auto_parallel::str_join;
 
-std::string GetBroadcastAxes(const int64_t& tenosr_ndim,
+std::string GetBroadcastAxes(const int64_t& tensor_ndim,
                              const int64_t& broadcast_ndim,
                              const std::string& alphabet) {
   PADDLE_ENFORCE_GE(
@@ -38,15 +38,15 @@ std::string GetBroadcastAxes(const int64_t& tenosr_ndim,
           alphabet.size(),
           broadcast_ndim));
   PADDLE_ENFORCE_GE(broadcast_ndim,
-                    tenosr_ndim,
+                    tensor_ndim,
                     phi::errors::InvalidArgument(
-                        "The broadcast ndim [%d] is less than tenosr ndim [%d]",
+                        "The broadcast ndim [%d] is less than tensor ndim [%d]",
                         broadcast_ndim,
-                        tenosr_ndim));
-  if (tenosr_ndim <= 0) {
+                        tensor_ndim));
+  if (tensor_ndim <= 0) {
     return std::string();
   }
-  return alphabet.substr(broadcast_ndim - tenosr_ndim, tenosr_ndim);
+  return alphabet.substr(broadcast_ndim - tensor_ndim, tensor_ndim);
 }
 
 // Rule1: A repicated dimension could be merged by any sharded dimension.

paddle/phi/infermeta/spmd_rules/utils.h

@@ -33,11 +33,11 @@ inline bool IsEmpty(const std::vector<int64_t>& shape) {
 }
 
 // Generate the axis notation of tensor for the einsum notation of a broadcast
-// operation(alignment star from the rightmost axis). tenosr_ndim: the size of
+// operation(alignment star from the rightmost axis). tensor_ndim: the size of
 // the tensor. broadcast_ndim: the maxium size of tensors in this broadcast
 // operation. alphabet: the characters used to represent the axes of tensor.
 // length of alphabet should >= broadcast_ndim.
-std::string GetBroadcastAxes(const int64_t& tenosr_ndim,
+std::string GetBroadcastAxes(const int64_t& tensor_ndim,
                              const int64_t& broadcast_ndim,
                              const std::string& alphabet);
 

python/paddle/static/nn/sequence_lod.py

@@ -314,7 +314,7 @@ def sequence_pool(input, pool_type, is_test=False, pad_value=0.0):
     Args:
         input (variable): Tensor with lod_level no more than 2. The data type should be float32 or float64.
         pool_type (str): The pooling type that supports average, sum, sqrt, max, last or first.
-        is_test (bool): Only works when :attr:`pool_type` is max. If set False, a temporary Tenosr maxIndex is
+        is_test (bool): Only works when :attr:`pool_type` is max. If set False, a temporary Tensor maxIndex is
             created to record the index information corresponding to the maximum value, which is used for backward
             gradient calculation in the training phase. Default: False.
         pad_value (float): Used to pad the pooling result for empty input sequence. Default: 0.0

python/paddle/tensor/creation.py

@@ -2841,7 +2841,7 @@ def cauchy_(x, loc=0, scale=1, name=None):
     """Fills the tensor with numbers drawn from the Cauchy distribution.
 
     Args:
-        x (Tenosr): the tensor will be filled, The data type is float32 or float64.
+        x (Tensor): the tensor will be filled, The data type is float32 or float64.
         loc (scalar, optional):  Location of the peak of the distribution. The data type is float32 or float64.
         scale (scalar, optional): The half-width at half-maximum (HWHM). The data type is float32 or float64. Must be positive values.
         name (str, optional): For details, please refer to :ref:`api_guide_Name`. Generally, no setting is required. Default: None.
@@ -2875,7 +2875,7 @@ def geometric_(x, probs, name=None):
     """Fills the tensor with numbers drawn from the Geometric distribution.
 
     Args:
-        x (Tenosr): the tensor will be filled, The data type is float32 or float64.
+        x (Tensor): the tensor will be filled, The data type is float32 or float64.
         probs (Real|Tensor): Probability parameter.
             The value of probs must be positive. When the parameter is a tensor, probs is probability of success for each trial.
         name (str, optional): For details, please refer to :ref:`api_guide_Name`. Generally, no setting is required. Default: None.