Change LoDTensorDesc to DenseTensorDesc (#1446)

* Fix

* Fix

* Fix

---------

Co-authored-by: Zheng-Bicheng <[email protected]>

paddle2onnx/parser/parser.cc

@@ -354,7 +354,7 @@ const framework::proto::OpDesc& PaddleParser::GetOpDesc(int32_t block_idx,
 }
 
 void PaddleParser::InitBlock() {
-  //  if (ExistsDumplicateTensorName()) {
+  //  if (ExistsDuplicateTensorName()) {
   //    return false;
   //  }
   GetBlocksVarName2Id();
@@ -485,7 +485,7 @@ TensorInfo PaddleParser::GetTensorInfo(
     return info;
   }
 
-  auto tensor = prog->blocks(block_idx).vars(var_idx).type().lod_tensor();
+  auto tensor = prog->blocks(block_idx).vars(var_idx).type().dense_tensor();
   TensorInfo info;
   info.name = name;
   info.dtype = tensor.tensor().data_type();

paddle2onnx/proto/p2o_paddle.proto

@@ -194,19 +194,19 @@ message VarType {
   }
   optional TensorDesc selected_rows = 2;
 
-  message LoDTensorDesc {
+  message DenseTensorDesc {
     required TensorDesc tensor = 1;
     optional int32 lod_level = 2 [ default = 0 ];
   }
-  optional LoDTensorDesc lod_tensor = 3;
+  optional DenseTensorDesc dense_tensor = 3;
 
-  message LoDTensorArrayDesc {
+  message DenseTensorArrayDesc {
     required TensorDesc tensor = 1;
     optional int32 lod_level = 2 [ default = 0 ];
   }
-  optional LoDTensorArrayDesc tensor_array = 4;
+  optional DenseTensorArrayDesc tensor_array = 4;
 
-  message ReaderDesc { repeated LoDTensorDesc lod_tensor = 1; }
+  message ReaderDesc { repeated DenseTensorDesc dense_tensor = 1; }
   optional ReaderDesc reader = 5;
 
   message Tuple { repeated Type element_type = 1; }

tests/detection_ops/nms.py

@@ -37,7 +37,7 @@ def multiclass_nms(bboxes,
                            N is the batch size. Each bounding box has four
                            coordinate values and the layout is
                            [xmin, ymin, xmax, ymax], when box size equals to 4.
-                           2. (LoDTensor) A 3-D Tensor with shape [M, C, 4]
+                           2. (DenseTensor) A 3-D Tensor with shape [M, C, 4]
                            M is the number of bounding boxes, C is the
                            class number
         scores (Tensor): Two types of scores are supported:
@@ -48,7 +48,7 @@ class number
                            are total M scores which corresponding M bounding
                            boxes. Please note, M is equal to the 2nd dimension
                            of BBoxes.
-                           2. (LoDTensor) A 2-D LoDTensor with shape [M, C].
+                           2. (DenseTensor) A 2-D DenseTensor with shape [M, C].
                            M is the number of bbox, C is the class number.
                            In this case, input BBoxes should be the second
                            case with shape [M, C, 4].
@@ -76,14 +76,14 @@ class number
     Returns:
         A tuple with two Variables: (Out, Index) if return_index is True,
         otherwise, a tuple with one Variable(Out) is returned.
-        Out: A 2-D LoDTensor with shape [No, 6] represents the detections.
+        Out: A 2-D DenseTensor with shape [No, 6] represents the detections.
         Each row has 6 values: [label, confidence, xmin, ymin, xmax, ymax]
-        or A 2-D LoDTensor with shape [No, 10] represents the detections.
+        or A 2-D DenseTensor with shape [No, 10] represents the detections.
         Each row has 10 values: [label, confidence, x1, y1, x2, y2, x3, y3,
         x4, y4]. No is the total number of detections.
         If all images have not detected results, all elements in LoD will be
         0, and output tensor is empty (None).
-        Index: Only return when return_index is True. A 2-D LoDTensor with
+        Index: Only return when return_index is True. A 2-D DenseTensor with
         shape [No, 1] represents the selected index which type is Integer.
         The index is the absolute value cross batches. No is the same number
         as Out. If the index is used to gather other attribute such as age,

tests/test_auto_scan_distribute_fpn_proposals_v2.py

@@ -40,7 +40,7 @@ def distribute_fpn_proposals(fpn_rois,
                              name=None):
     r"""
 
-    **This op only takes LoDTensor as input.** In Feature Pyramid Networks
+    **This op only takes DenseTensor as input.** In Feature Pyramid Networks
     (FPN) models, it is needed to distribute all proposals into different FPN
     level, with respect to scale of the proposals, the referring scale and the
     referring level. Besides, to restore the order of proposals, we return an
@@ -77,7 +77,7 @@ def distribute_fpn_proposals(fpn_rois,
     Returns:
         Tuple:
 
-        multi_rois(List) : A list of 2-D LoDTensor with shape [M, 4]
+        multi_rois(List) : A list of 2-D DenseTensor with shape [M, 4]
         and data type of float32 and float64. The length is
         max_level-min_level+1. The proposals in each FPN level.
 

tests/test_auto_scan_roi_align.py

@@ -60,7 +60,7 @@ def roi_align(input,
             where N is the batch size, C is the input channel, H is Height, W is weight.
             The data type is float32 or float64.
         rois (Tensor): ROIs (Regions of Interest) to pool over.It should be
-            a 2-D Tensor or 2-D LoDTensor of shape (num_rois, 4), the lod level is 1.
+            a 2-D Tensor or 2-D DenseTensor of shape (num_rois, 4), the lod level is 1.
             The data type is float32 or float64. Given as [[x1, y1, x2, y2], ...],
             (x1, y1) is the top left coordinates, and (x2, y2) is the bottom right coordinates.
         output_size (int or tuple[int, int]): The pooled output size(h, w), data type is int32. If int, h and w are both equal to output_size.