add gpu kernel by copying inputs/outputs between cpu and gpu.

paddle/framework/operator.cc

@@ -38,7 +38,7 @@ const Tensor* GetTensorFromVar(const Variable* var) {
     return &var->Get<LoDTensor>();
   }
   PADDLE_ENFORCE(var->IsType<Tensor>(),
-                 "The Input must be LoDTensor or Tensor.");
+                 "The Input must be a LoDTensor or a Tensor.");
   return &var->Get<Tensor>();
 }
 
@@ -47,39 +47,39 @@ Tensor* GetTensorFromVar(Variable* var) {
     return var->GetMutable<LoDTensor>();
   }
   PADDLE_ENFORCE(var->IsType<Tensor>(),
-                 "The Input must be LoDTensor or Tensor.");
+                 "The Input must be a LoDTensor or a Tensor.");
   return var->GetMutable<Tensor>();
 }
 
 std::string OperatorBase::Input(const std::string& name) const {
   auto& ins = Inputs(name);
   PADDLE_ENFORCE_LE(ins.size(), 1UL,
-                    "Op %s input %s should contain only one variable", type_,
-                    name);
+                    "Operator %s's input %s should contain only one variable.",
+                    type_, name);
   return ins.empty() ? kEmptyVarName : ins[0];
 }
 
 const std::vector<std::string>& OperatorBase::Inputs(
     const std::string& name) const {
   auto it = inputs_.find(name);
-  PADDLE_ENFORCE(it != inputs_.end(), "Op %s do not have input %s", type_,
-                 name);
+  PADDLE_ENFORCE(it != inputs_.end(), "Operator %s does not have the input %s.",
+                 type_, name);
   return it->second;
 }
 
 std::string OperatorBase::Output(const std::string& name) const {
   auto& outs = Outputs(name);
   PADDLE_ENFORCE_LE(outs.size(), 1UL,
-                    "Op %s output %s should contain only one variable", type_,
-                    name);
+                    "Operator %s's output %s should contain only one variable.",
+                    type_, name);
   return outs.empty() ? kEmptyVarName : outs[0];
 }
 
 const std::vector<std::string>& OperatorBase::Outputs(
     const std::string& name) const {
   auto it = outputs_.find(name);
-  PADDLE_ENFORCE(it != outputs_.end(), "Op %s does not have output called %s",
-                 type_, name);
+  PADDLE_ENFORCE(it != outputs_.end(),
+                 "Operator %s does not have an output called %s.", type_, name);
   return it->second;
 }
 

paddle/framework/tensor_impl.h

@@ -108,9 +108,10 @@ inline void* Tensor::mutable_data(platform::Place place, std::type_index type) {
   if (holder_ != nullptr) {
     holder_->set_type(type);
   }
-  PADDLE_ENFORCE_GT(numel(), 0,
-                    "Tensor's numel must be larger than zero to call "
-                    "Tensor::mutable_data. Call Tensor::set_dim first.");
+  PADDLE_ENFORCE_GT(
+      numel(), 0,
+      "When calling this method, the Tensor's numel must be larger than zero. "
+      "Please check Tensor::Resize has been called first.");
   int64_t size = numel() * SizeOfType(type);
   /* some versions of boost::variant don't have operator!= */
   if (holder_ == nullptr || !(holder_->place() == place) ||

paddle/operators/linear_chain_crf_op.cc

@@ -204,8 +204,7 @@ class LinearChainCRFGradOp : public framework::OperatorWithKernel {
     PADDLE_ENFORCE(emission_exps_dims[0],
                    "An empty mini-batch is not allowed.");
 
-    auto transition_exps_dims =
-        ctx->GetInputDim(framework::GradVarName("TransitionExps"));
+    auto transition_exps_dims = ctx->GetInputDim("TransitionExps");
     PADDLE_ENFORCE_EQ(transition_exps_dims.size(), 2UL,
                       "The Input(TransitionExps) should be a 2-D tensor.");
     PADDLE_ENFORCE_EQ(
@@ -240,7 +239,8 @@ class LinearChainCRFGradOp : public framework::OperatorWithKernel {
   // operator is determined by its input: graidents of LogLikelihood.
   framework::DataType IndicateDataType(
       const framework::ExecutionContext& ctx) const override {
-    return framework::ToDataType(ctx.Input<LoDTensor>("LogLikelihood")->type());
+    return framework::ToDataType(
+        ctx.Input<LoDTensor>(framework::GradVarName("LogLikelihood"))->type());
   }
 };
 

paddle/operators/linear_chain_crf_op.cu

@@ -0,0 +1,26 @@
+/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License. */
+
+#include "paddle/operators/linear_chain_crf_op.h"
+
+namespace ops = paddle::operators;
+
+REGISTER_OP_GPU_KERNEL(
+    linear_chain_crf,
+    ops::LinearChainCRFOpKernel<paddle::platform::GPUPlace, float>,
+    ops::LinearChainCRFOpKernel<paddle::platform::GPUPlace, double>);
+REGISTER_OP_GPU_KERNEL(
+    linear_chain_crf_grad,
+    ops::LinearChainCRFGradOpKernel<paddle::platform::GPUPlace, float>,
+    ops::LinearChainCRFGradOpKernel<paddle::platform::GPUPlace, double>);