From dc520da763f258e6ca1c79f79bb82bc55ec4cf14 Mon Sep 17 00:00:00 2001 From: caoying03 Date: Tue, 5 Sep 2017 23:08:21 +0800 Subject: [PATCH] update doc of softmax_op. --- paddle/operators/softmax_op.cc | 26 ++++++++++++++++++++++---- 1 file changed, 22 insertions(+), 4 deletions(-) diff --git a/paddle/operators/softmax_op.cc b/paddle/operators/softmax_op.cc index 40c51a64c4..7d062ad67c 100644 --- a/paddle/operators/softmax_op.cc +++ b/paddle/operators/softmax_op.cc @@ -24,7 +24,7 @@ class SoftmaxOp : public framework::OperatorWithKernel { protected: void InferShape(const framework::InferShapeContext &ctx) const override { PADDLE_ENFORCE(ctx.Input("X")->dims().size() == 2UL, - "The input of softmax op must be matrix"); + "The input of softmax op must be a matrix."); ctx.Output("Y")->Resize(ctx.Input("X")->dims()); } }; @@ -34,9 +34,27 @@ class SoftmaxOpMaker : public framework::OpProtoAndCheckerMaker { SoftmaxOpMaker(framework::OpProto *proto, framework::OpAttrChecker *op_checker) : OpProtoAndCheckerMaker(proto, op_checker) { - AddInput("X", "input of softmax"); - AddOutput("Y", "output of softmax"); - AddComment("Softmax Op"); + AddInput("X", + "The input tensor of softmax. " + "2-D with shape [batch_size, input_feature_dimensions]."); + AddOutput("Y", "The normalized values with the same shape as X."); + AddComment(R"DOC( +The input of softmax operator is a 2-D tensor with shape N x K (N is the +batch_size, K is the dimension of input feature). The output tensor has the +same shape as the input tensor. + +For each row of the input tensor, the softmax operator squashes the +K-dimensional vector of arbitrary real values to a K-dimensional vector of real +values in the range [0, 1] that add up to 1. Specifically, it computes the +exponential of the given dimension and the sum of exponential values of all +the other dimensions in the K-dimensional vector input. Then the ratio of the +exponential of the given dimension and the sum of exponential values of all +the other dimensions is the output of the softmax operator. + +For each row `i` and each column `j` in X, we have: + Y[i, j] = exp(X[i, j]) / sum_j(exp(X[i, j])) + +)DOC"); } }; -- GitLab