From: Shubham Gupta/SNAP /SRI-Bangalore/Engineer/삼성전자 Date: Mon, 7 Jan 2019 02:16:30 +0000 (+0530) Subject: Correcting assertion in transposeConv CPU implementation (#4153) X-Git-Tag: 0.3~7 X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=720264917e9637e417746236fee672c215a4df3f;p=platform%2Fcore%2Fml%2Fnnfw.git Correcting assertion in transposeConv CPU implementation (#4153) This patch will correct the position of bracket which has been wrongly placed. Signed-off-by: shubham --- diff --git a/runtimes/pure_arm_compute/src/internal/layers/SimpleTransposeConv.cc b/runtimes/pure_arm_compute/src/internal/layers/SimpleTransposeConv.cc index ee6c806..abc2912 100644 --- a/runtimes/pure_arm_compute/src/internal/layers/SimpleTransposeConv.cc +++ b/runtimes/pure_arm_compute/src/internal/layers/SimpleTransposeConv.cc @@ -58,7 +58,7 @@ inline void ApplyTransposeConv( assert(batches == output_shape[axises[0]]); assert(input_depth == filter_shape[axises[3]]); - assert(filter_shape[axises[0] == output_depth]); + assert(filter_shape[axises[0]] == output_depth); // Although transpose convolution simplifies to convolution with transposed // weights for strides of 1, non-unitary striding complicates matters. To