Implement getOutputSize and getInputSize for seq layers

This commit is contained in:
2024-04-22 20:31:58 +02:00
parent fbf6c44bdd
commit f17debc244
18 changed files with 186 additions and 66 deletions

View File

@@ -187,6 +187,8 @@ TEST(MatMulTest, SumReduceTest) {
cudaMemcpy(d_input, input.data(), sizeof(float) * n, cudaMemcpyHostToDevice);
EXPECT_EQ(cudaStatus, cudaSuccess);
CUDANet::Utils::clear(d_sum, n);
CUDANet::Kernels::sum_reduce<<<gridSize, BLOCK_SIZE>>>(
d_input, d_sum, n
);
@@ -208,7 +210,5 @@ TEST(MatMulTest, SumReduceTest) {
EXPECT_FLOAT_EQ(expected, sum[0]);
cudaFree(d_input);
cudaFree(d_sum);
cudaFree(d_sum);
}