Refine the doc for mul_op and fully connected layer. #6792

lcy-seso · 2017-12-20T09:42:42Z

abhinavarora

Minor grammatical changes. Make sure to follow the comments syntax in the naming convention for inputs and outputs.

abhinavarora · 2017-12-20T18:55:41Z

paddle/operators/mul_op.cc

-    AddInput("X", "The first input of mul op");
-    AddInput("Y", "The second input of mul op");
-    AddOutput("Out", "The output of mul op");
+    AddInput("X", "The first input tensor of the mul op.");


This should The first input tensor of mul op. There should not be a the

abhinavarora · 2017-12-20T18:55:48Z

paddle/operators/mul_op.cc

-    AddInput("Y", "The second input of mul op");
-    AddOutput("Out", "The output of mul op");
+    AddInput("X", "The first input tensor of the mul op.");
+    AddInput("Y", "The second input tensor of the mul op.");


Same as above

abhinavarora · 2017-12-20T18:56:05Z

paddle/operators/mul_op.cc

-    AddOutput("Out", "The output of mul op");
+    AddInput("X", "The first input tensor of the mul op.");
+    AddInput("Y", "The second input tensor of the mul op.");
+    AddOutput("Out", "The output tensor of the mul op.");


Same as above.

abhinavarora · 2017-12-20T18:59:37Z

paddle/operators/mul_op.cc

@@ -73,30 +73,42 @@ class MulOpMaker : public framework::OpProtoAndCheckerMaker {
 public:
  MulOpMaker(OpProto* proto, OpAttrChecker* op_checker)
      : OpProtoAndCheckerMaker(proto, op_checker) {


We should also be following the comment syntax as given in the naming convention -> https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/operators/name_convention.md

I check the name convention in this doc https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/operators/op_documentation/name_convention.md, the current input names and the output name do not violate our name convention. So I only fix the comment style suggested in this doc. If I miss still something, please point it out. Thank you.

abhinavarora · 2017-12-20T19:01:17Z

paddle/operators/mul_op.cc

+              flattened to form the first dimension of the final matrix (height
+              of the matrix), and the rest `rank(X) - num_col_dims` dimensions
+              are flattened to form the second dimension of the final matrix (
+              width of the matrix). As a result, height of the flattened matrix


height -> the height

abhinavarora · 2017-12-20T19:02:05Z

paddle/operators/mul_op.cc

+              are flattened to form the second dimension of the final matrix (
+              width of the matrix). As a result, height of the flattened matrix
+              is equal to the product of `X`'s first `x_num_col_dims` dimensions'
+              sizes, and width of the flattened matrix is equal to the product


width -> the width

abhinavarora · 2017-12-20T19:02:28Z

paddle/operators/mul_op.cc

+              sizes, and width of the flattened matrix is equal to the product
+              of `X`'s last `rank(x) - num_col_dims` dimensions' size.
+              For example, suppose `X` is a 6-dimensional tensor with the shape
+              [2, 3, 4, 5, 6], and `x_num_col_dims` = 3. Then, the flattened


Then -> Thus

abhinavarora · 2017-12-20T19:02:52Z

paddle/operators/mul_op.cc

-             in that case, tensors will be reshaped to a matrix. Just like input `X`.
+        R"DOC(The mul_op can take tensors with more than two dimensions as its
+              inputs. If the input `Y` is a tensor with more than two
+              dimensions, `Y` will be flatten into a two-dimensional matrix


flatten -> flattened

abhinavarora · 2017-12-20T19:03:30Z

python/paddle/v2/fluid/layers/nn.py

+                         tensor will first be flattened into a 2-dimensional
+                         matrix. The parameter `num_flatten_dims` determines
+                         how the input tensor is flattened: the first
+                         `num_flatten_dims` dimensions will be flatten to form


flatten -> flattened

lcy-seso

follow comments.

lcy-seso · 2017-12-21T07:20:47Z

paddle/operators/mul_op.cc

-    AddInput("X", "The first input of mul op");
-    AddInput("Y", "The second input of mul op");
-    AddOutput("Out", "The output of mul op");
+    AddInput("X", "The first input tensor of the mul op.");


lcy-seso · 2017-12-21T07:20:58Z

paddle/operators/mul_op.cc

-    AddInput("Y", "The second input of mul op");
-    AddOutput("Out", "The output of mul op");
+    AddInput("X", "The first input tensor of the mul op.");
+    AddInput("Y", "The second input tensor of the mul op.");


lcy-seso · 2017-12-21T07:21:04Z

paddle/operators/mul_op.cc

-    AddOutput("Out", "The output of mul op");
+    AddInput("X", "The first input tensor of the mul op.");
+    AddInput("Y", "The second input tensor of the mul op.");
+    AddOutput("Out", "The output tensor of the mul op.");


lcy-seso · 2017-12-21T07:22:41Z

paddle/operators/mul_op.cc

+              flattened to form the first dimension of the final matrix (height
+              of the matrix), and the rest `rank(X) - num_col_dims` dimensions
+              are flattened to form the second dimension of the final matrix (
+              width of the matrix). As a result, height of the flattened matrix


lcy-seso · 2017-12-21T07:22:47Z

paddle/operators/mul_op.cc

+              are flattened to form the second dimension of the final matrix (
+              width of the matrix). As a result, height of the flattened matrix
+              is equal to the product of `X`'s first `x_num_col_dims` dimensions'
+              sizes, and width of the flattened matrix is equal to the product


lcy-seso · 2017-12-21T07:25:31Z

paddle/operators/mul_op.cc

-             in that case, tensors will be reshaped to a matrix. Just like input `X`.
+        R"DOC(The mul_op can take tensors with more than two dimensions as its
+              inputs. If the input `Y` is a tensor with more than two
+              dimensions, `Y` will be flatten into a two-dimensional matrix


lcy-seso · 2017-12-21T07:25:51Z

paddle/operators/mul_op.cc

+              sizes, and width of the flattened matrix is equal to the product
+              of `X`'s last `rank(x) - num_col_dims` dimensions' size.
+              For example, suppose `X` is a 6-dimensional tensor with the shape
+              [2, 3, 4, 5, 6], and `x_num_col_dims` = 3. Then, the flattened


lcy-seso · 2017-12-21T11:04:16Z

paddle/operators/mul_op.cc

@@ -73,30 +73,42 @@ class MulOpMaker : public framework::OpProtoAndCheckerMaker {
 public:
  MulOpMaker(OpProto* proto, OpAttrChecker* op_checker)
      : OpProtoAndCheckerMaker(proto, op_checker) {


I check the name convention in this doc https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/operators/op_documentation/name_convention.md, the current input names and the output name do not violate our name convention. So I only fix the comment style suggested in this doc. If I miss still something, please point it out. Thank you.

abhinavarora · 2017-12-21T19:11:26Z

LGTM!

lcy-seso added 2 commits December 20, 2017 17:20

refine the doc.

6a1e312

Merge branch 'develop' into refine_doc

6646873

lcy-seso requested review from abhinavarora and pkuyym December 20, 2017 09:46

some small refines.

c322c7b

lcy-seso force-pushed the refine_doc branch from b15e65d to c322c7b Compare December 20, 2017 09:59

abhinavarora suggested changes Dec 20, 2017

View reviewed changes

Merge branch 'develop' into refine_doc

ebe4425

lcy-seso force-pushed the refine_doc branch from fee06cd to f2db1b8 Compare December 21, 2017 11:02

lcy-seso commented Dec 21, 2017

View reviewed changes

follow comments.

a74db48

lcy-seso force-pushed the refine_doc branch from f2db1b8 to a74db48 Compare December 21, 2017 12:41

abhinavarora approved these changes Dec 21, 2017

View reviewed changes

lcy-seso merged commit 298dc89 into PaddlePaddle:develop Dec 22, 2017

lcy-seso deleted the refine_doc branch December 23, 2017 09:30

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Refine the doc for mul_op and fully connected layer. #6792

Refine the doc for mul_op and fully connected layer. #6792

lcy-seso commented Dec 20, 2017 •

edited

Loading

abhinavarora left a comment

abhinavarora Dec 20, 2017

lcy-seso Dec 21, 2017

abhinavarora Dec 20, 2017

lcy-seso Dec 21, 2017

abhinavarora Dec 20, 2017

lcy-seso Dec 21, 2017

abhinavarora Dec 20, 2017

lcy-seso Dec 21, 2017 •

edited

Loading

abhinavarora Dec 20, 2017

lcy-seso Dec 21, 2017

abhinavarora Dec 20, 2017

lcy-seso Dec 21, 2017

abhinavarora Dec 20, 2017

lcy-seso Dec 21, 2017

abhinavarora Dec 20, 2017

lcy-seso Dec 21, 2017

abhinavarora Dec 20, 2017

lcy-seso left a comment

lcy-seso Dec 21, 2017

lcy-seso Dec 21, 2017

lcy-seso Dec 21, 2017

lcy-seso Dec 21, 2017

lcy-seso Dec 21, 2017

lcy-seso Dec 21, 2017

lcy-seso Dec 21, 2017

lcy-seso Dec 21, 2017 •

edited

Loading

abhinavarora commented Dec 21, 2017

Refine the doc for mul_op and fully connected layer. #6792

Refine the doc for mul_op and fully connected layer. #6792

Conversation

lcy-seso commented Dec 20, 2017 • edited Loading

abhinavarora left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

lcy-seso Dec 21, 2017 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

lcy-seso left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

lcy-seso Dec 21, 2017 • edited Loading

Choose a reason for hiding this comment

abhinavarora commented Dec 21, 2017

lcy-seso commented Dec 20, 2017 •

edited

Loading

lcy-seso Dec 21, 2017 •

edited

Loading

lcy-seso Dec 21, 2017 •

edited

Loading