apeforest commented on a change in pull request #14031: Fix transposed convolution in CPU w/o MKLDNN. URL: https://github.com/apache/incubator-mxnet/pull/14031#discussion_r262752392
########## File path: src/operator/nn/deconvolution-inl.h ########## @@ -373,55 +357,46 @@ class DeconvolutionOp { in_data[deconv::kWeight].get_with_shape<xpu, 3, DType>(wmat_shape, s); Tensor<xpu, 3, DType> gwmat = in_grad[deconv::kWeight].get_with_shape<xpu, 3, DType>(wmat_shape, s); -#if defined(__CUDACC__) - CHECK_EQ(s->blas_handle_ownership_, Stream<xpu>::OwnHandle) - << "Must init CuBLAS handle in stream"; -#endif const index_t nbatch = data.size(0); Tensor<xpu, 1, DType> workspace = ctx.requested[deconv::kTempSpace].get_space_typed<xpu, 1, DType>( Shape1(this->InitTemp(grad.shape_, data.shape_)), s); - for (index_t i = 0; i < nbatch; i += nstep_) { - const index_t step = std::min(nstep_, nbatch - i); Review comment: I think it's used to convert multiple batch of image data into columns in the prevous library. However, it is not supported in the `col2im` method. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services