Mojo function

conv_gpu

conv_gpu[conv_rank: Int, //, input_type: DType, filter_type: DType, output_type: DType, maybe_epilogue_func: Optional[def[dtype: DType, rank: Int, width: Int](IndexList[rank], SIMD[dtype, width]) capturing -> None] = None, filter_is_fcrs: Bool = False, has_residual: Bool = False](input: TileTensor[input_type, input.LayoutType, input.origin, linear_idx_type=input.linear_idx_type, element_size=input.element_size], filter: TileTensor[filter_type, filter.LayoutType, filter.origin, linear_idx_type=filter.linear_idx_type, element_size=filter.element_size], output: TileTensor[output_type, output.LayoutType, output.origin, linear_idx_type=output.linear_idx_type, element_size=output.element_size], stride: IndexList[conv_rank], dilation: IndexList[conv_rank], padding: IndexList[(2 * conv_rank)], num_groups: Int, ctx: DeviceContext, source_ptr: UnsafePointer[Scalar[output_type], MutAnyOrigin] = UnsafePointer(_unsafe_null=Tuple()), beta: Float32 = 0)