When we use MNNConverter to convert other mobilenet model to MNN model, {Conv3x3Depthwise + BN + Relu + Conv1x1 + BN + Relu} will be converted and optimized to {Conv3x3Depthwise + Conv1x1}
| 25 | // {Conv3x3Depthwise + BN + Relu + Conv1x1 + BN + Relu} will be converted |
| 26 | // and optimized to {Conv3x3Depthwise + Conv1x1} |
| 27 | static VARP convBlock(VARP x, INTS channels, int stride) { |
| 28 | int inputChannel = channels[0], outputChannel = channels[1]; |
| 29 | int group = inputChannel; |
| 30 | x = _Conv(0.01f, 0.0f, x, {inputChannel, inputChannel}, {3, 3}, SAME, {stride, stride}, {1, 1}, group); |
| 31 | x = _Conv(0.03f, 0.0f, x, {inputChannel, outputChannel}, {1, 1}, SAME, {1, 1}, {1, 1}, 1); |
| 32 | return x; |
| 33 | } |
| 34 | static VARP convBlocTemp(VARP x, INTS channels, int stride) { |
| 35 | int inputChannel = channels[0], outputChannel = channels[1]; |
| 36 | int group = inputChannel; |
no test coverage detected