add dropout:"add dropout to develop-move-mlu-pass-forward branch for run senet101"

lite/kernels/mlu/bridges/dropout_op.cc

@@ -33,10 +33,15 @@ int DropoutConverter(void* ctx, OpLite* op, KernelBase* kernel) {
   // Create act node and set params from op
   auto x_var_name = op_info->Input("X").front();
   auto out_var_name = op_info->Output("Out").front();
+  auto mask_var_name = op_info->Output("Mask").front();
   auto output = scope->FindVar(out_var_name)->GetMutable<Tensor>();
   auto output_dims = output->dims().Vectorize();
   auto output_tensor = graph->AddNode(
       out_var_name, output_dims, CNML_TENSOR, CNML_NCHW, graph->FPType());
+  auto mask = scope->FindVar(mask_var_name)->GetMutable<Tensor>();
+  auto mask_dims = mask->dims().Vectorize();
+  auto mask_tensor = graph->AddNode(
+      mask_var_name, mask_dims, CNML_TENSOR, CNML_NCHW, graph->FPType());
 
   // is_test is true by default
   // if(op_info->HasAttr("is_test")){

lite/kernels/mlu/bridges/dropout_op_test.cc

@@ -85,9 +85,9 @@ void test_dropout(int bs,
   opdesc.SetAttr("seed", seed);
   opdesc.SetAttr("dropout_implementation", dropout_implementation);
   opdesc.SetAttr("dropout_prob", dropout_prob);
+  VLOG(6) << "mask: " << mask->dims()[0] << std::endl;
   // create and convert op to MLU model, then run it on MLU
   auto op = CreateOp<operators::DropoutOp>(opdesc, &scope);
-  VLOG(6) << "mask: " << mask << std::endl;
   dropout_ref(op);
   out_ref->CopyDataFrom(*out);