Bug fix

paddle/function/ConvOp.h

@@ -58,7 +58,7 @@ public:
     CHECK_EQ(outputs[0].shape().ndims(), (size_t)4);
 
     CHECK(inputs[0].shape()[0] == outputs[0].shape()[0]);
-    CHECK(inputs[0].shape()[1] == inputs[1].shape()[1]);
+    CHECK(inputs[0].shape()[1] / groups_ == inputs[1].shape()[1]);
     CHECK(outputs[0].shape()[1] == inputs[1].shape()[0]);
   }
 

paddle/function/ConvOpTest.cpp

@@ -83,9 +83,11 @@ TEST(Convolution, GEMM) {
                                                          "GemmConv-CPU");
 }
 
+#ifndef PADDLE_ONLY_CPU
 TEST(Convolution, GEMM2) {
   ConvolutionTest<DEVICE_TYPE_CPU, DEVICE_TYPE_GPU> test("GemmConv-CPU",
                                                          "GemmConv-GPU");
 }
+#endif
 
 }  // namespace paddle

paddle/function/GemmConvOp.cpp

@@ -101,8 +101,6 @@ public:
     size_t outputHeight = outputs[0].shape()[2];
     size_t outputWidth = outputs[0].shape()[3];
 
-    CHECK_EQ(inputChannels / groups_, inputs[1].shape()[1]);
-
     real* inputData = inputs[0].data<real>();
     real* filterData = inputs[1].data<real>();
     real* outputData = outputs[0].data<real>();
@@ -134,9 +132,9 @@ public:
                outputWidth,
                colData);
 
-        int M = outputChannels;
+        int M = outputChannels / groups_;
         int N = outputHeight * outputWidth;
-        int K = inputChannels * filterHeight * filterWidth;
+        int K = inputChannels / groups_ * filterHeight * filterWidth;
         gemm(M,
              N,
              K,