|
- # 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.
-
- import numpy as np
- from tests.common.tensorio import compare_tensor
- from akg.utils import kernel_exec as utils
- from tests.common.test_op.diagpart import diagpart
- from tests.common.gen_random import random_gaussian
-
- def diagpart_run(shape, dtype, kernel_name, attrs, cce_path="./"):
- def gen_data(shape, dtype):
- data = random_gaussian(shape, miu=1, sigma=0.1).astype(dtype)
- out_shape = []
- rank = len(shape)
- for i in range(rank // 2):
- out_shape.append(shape[i])
- expect = np.full(out_shape, 0.0, dtype)
- if rank == 2:
- for i in range(shape[0]):
- expect[i] = data[i, i]
- elif rank == 4:
- for i in range(shape[0]):
- for j in range(shape[1]):
- expect[i, j] = data[i, j, i, j]
- elif rank == 6:
- for i in range(shape[0]):
- for j in range(shape[1]):
- for m in range(shape[2]):
- expect[i, j, m] = data[i, j, m, i, j, m]
- elif rank == 8:
- for i in range(shape[0]):
- for j in range(shape[1]):
- for m in range(shape[2]):
- for n in range(shape[3]):
- expect[i, j, m, n] = data[i, j, m, n, i, j, m, n]
- else:
- raise RuntimeError("diagpart only support even rank (2/4/6/8) while the rank is {}".format(rank))
- return data, expect
-
- if 'tuning' in attrs.keys():
- t = attrs.get("tuning", False)
- kernel_name = attrs.get("kernel_name", False)
- testdata, expect = gen_data(shape, dtype)
- mod = utils.op_build_test(diagpart, [shape], [dtype], kernel_name=kernel_name, attrs=attrs, tuning=t)
- if t:
- out_shape = expect.shape
- out = np.full(out_shape, np.nan, dtype)
- return mod, expect, (testdata, out)
- else:
- return mod
- else:
- testdata, expect = gen_data(shape, dtype)
- out_shape = expect.shape
-
- out = np.full(out_shape, np.nan, dtype)
- mod = utils.op_build_test(diagpart, [shape], [dtype], kernel_name=kernel_name, attrs=attrs)
- res = utils.mod_launch(mod, (testdata, out), expect=expect)
- cpr_res = compare_tensor(res, expect, rtol=5e-03, equal_nan=True)
- return testdata, res, expect, cpr_res
|
