# Copyright 2019 Huawei Technologies Co., Ltd # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. """operator dsl function:prelu""" import akg.tvm import akg.topi from akg.utils import validation_check as vc_util from akg.lang.cce import vmax, vmin, vadd, vmul, broadcast def prelu(A, w): """ brief Computes prelu value of a tensor. \f[ out = max(0, A) + min(0, wA) \f] Args: inputs akg.tvm.Tensor of type float16, float32 Returns: akg.tvm.Tensor of same type and shape as inputs """ # num_parameters=1, init=0.25 shape1 = [x.value for x in A.shape] dtype1 = A.dtype shape2 = [x.value for x in w.shape] dtype2 = w.dtype assert len(shape1) == 4, "only support 4-dim pooling" # NCHW assert len(shape2) == 1, "only support 1-dim a" assert (shape2[0] == shape1[1] or shape2[0] == 1), "there is only two values are legitimate: 1, or the number of channels at input. Default: 1" check_list = ["float16", "float32"] if not (dtype1.lower() in check_list and dtype2.lower() in check_list): raise RuntimeError("tile_cce only support %s while dtype is %s and %s" % (",".join(check_list), dtype1, dtype2)) vc_util.check_shape(shape1) vc_util.check_shape(shape2) w_reshape = akg.topi.reshape(w, (1, shape2[0], 1, 1)) w_broadcast = akg.topi.broadcast_to(w_reshape, shape1) const_zero = broadcast(akg.tvm.const(0, A.dtype), shape1, output_dtype=A.dtype) res = vadd(vmax(A, const_zero), vmul(vmin(A, const_zero), w_broadcast)) return res