2020-03-27 14:49:12 +08:00
|
|
|
#!/usr/bin/env python
|
2022-05-27 20:53:41 +08:00
|
|
|
# Copyright 2019-2022 Huawei Technologies Co., Ltd
|
2020-03-27 14:49:12 +08:00
|
|
|
#
|
|
|
|
|
# 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 os
|
|
|
|
|
import pytest
|
|
|
|
|
|
|
|
|
|
import mindspore.dataset as ds
|
|
|
|
|
from mindspore.mindrecord import FileWriter
|
|
|
|
|
|
|
|
|
|
|
2021-10-19 16:52:57 +08:00
|
|
|
def create_cv_mindrecord(file_name, files_num):
|
2020-03-27 14:49:12 +08:00
|
|
|
"""tutorial for cv dataset writer."""
|
2021-10-19 16:52:57 +08:00
|
|
|
if os.path.exists(file_name):
|
|
|
|
|
os.remove(file_name)
|
|
|
|
|
if os.path.exists("{}.db".format(file_name)):
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
|
|
|
|
writer = FileWriter(file_name, files_num)
|
2021-09-14 20:45:47 +08:00
|
|
|
cv_schema_json = {"file_name": {"type": "string"},
|
|
|
|
|
"label": {"type": "int32"}, "data": {"type": "bytes"}}
|
|
|
|
|
data = [{"file_name": "001.jpg", "label": 43,
|
|
|
|
|
"data": bytes('0xffsafdafda', encoding='utf-8')}]
|
2020-03-27 14:49:12 +08:00
|
|
|
writer.add_schema(cv_schema_json, "img_schema")
|
|
|
|
|
writer.add_index(["file_name", "label"])
|
|
|
|
|
writer.write_raw_data(data)
|
|
|
|
|
writer.commit()
|
|
|
|
|
|
|
|
|
|
|
2021-10-19 16:52:57 +08:00
|
|
|
def create_diff_schema_cv_mindrecord(file_name, files_num):
|
2020-05-08 16:11:59 +08:00
|
|
|
"""tutorial for cv dataset writer."""
|
2021-10-19 16:52:57 +08:00
|
|
|
if os.path.exists(file_name):
|
|
|
|
|
os.remove(file_name)
|
|
|
|
|
if os.path.exists("{}.db".format(file_name)):
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
|
|
|
|
writer = FileWriter(file_name, files_num)
|
2021-09-14 20:45:47 +08:00
|
|
|
cv_schema_json = {"file_name_1": {"type": "string"},
|
|
|
|
|
"label": {"type": "int32"}, "data": {"type": "bytes"}}
|
|
|
|
|
data = [{"file_name_1": "001.jpg", "label": 43,
|
|
|
|
|
"data": bytes('0xffsafdafda', encoding='utf-8')}]
|
2020-05-08 16:11:59 +08:00
|
|
|
writer.add_schema(cv_schema_json, "img_schema")
|
|
|
|
|
writer.add_index(["file_name_1", "label"])
|
|
|
|
|
writer.write_raw_data(data)
|
|
|
|
|
writer.commit()
|
|
|
|
|
|
2020-05-18 10:31:46 +08:00
|
|
|
|
2021-10-19 16:52:57 +08:00
|
|
|
def create_diff_page_size_cv_mindrecord(file_name, files_num):
|
2020-05-08 16:11:59 +08:00
|
|
|
"""tutorial for cv dataset writer."""
|
2021-10-19 16:52:57 +08:00
|
|
|
if os.path.exists(file_name):
|
|
|
|
|
os.remove(file_name)
|
|
|
|
|
if os.path.exists("{}.db".format(file_name)):
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
|
|
|
|
writer = FileWriter(file_name, files_num)
|
2020-05-18 10:31:46 +08:00
|
|
|
writer.set_page_size(1 << 26) # 64MB
|
2021-09-14 20:45:47 +08:00
|
|
|
cv_schema_json = {"file_name": {"type": "string"},
|
|
|
|
|
"label": {"type": "int32"}, "data": {"type": "bytes"}}
|
|
|
|
|
data = [{"file_name": "001.jpg", "label": 43,
|
|
|
|
|
"data": bytes('0xffsafdafda', encoding='utf-8')}]
|
2020-05-08 16:11:59 +08:00
|
|
|
writer.add_schema(cv_schema_json, "img_schema")
|
|
|
|
|
writer.add_index(["file_name", "label"])
|
|
|
|
|
writer.write_raw_data(data)
|
|
|
|
|
writer.commit()
|
|
|
|
|
|
2020-05-18 10:31:46 +08:00
|
|
|
|
2020-03-27 14:49:12 +08:00
|
|
|
def test_cv_lack_json():
|
2022-06-03 09:24:09 +08:00
|
|
|
"""
|
|
|
|
|
Feature: MindDataset
|
|
|
|
|
Description: Test MindDataset using json file that does not exist
|
|
|
|
|
Expectation: Exception is raised as expected
|
|
|
|
|
"""
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
2020-03-27 14:49:12 +08:00
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
2020-05-26 16:17:53 +08:00
|
|
|
with pytest.raises(Exception):
|
2021-10-19 16:52:57 +08:00
|
|
|
ds.MindDataset(file_name, "no_exist.json",
|
2021-09-14 20:45:47 +08:00
|
|
|
columns_list, num_readers)
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2020-03-27 14:49:12 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_cv_lack_mindrecord():
|
2022-06-03 09:24:09 +08:00
|
|
|
"""
|
|
|
|
|
Feature: MindDataset
|
|
|
|
|
Description: Test MindDataset using mindrecord that does not exist or no permission
|
|
|
|
|
Expectation: Exception is raised as expected
|
|
|
|
|
"""
|
2020-03-27 14:49:12 +08:00
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
|
|
|
|
with pytest.raises(Exception, match="does not exist or permission denied"):
|
2020-05-26 16:17:53 +08:00
|
|
|
_ = ds.MindDataset("no_exist.mindrecord", columns_list, num_readers)
|
2020-03-27 14:49:12 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_invalid_mindrecord():
|
2022-06-03 09:24:09 +08:00
|
|
|
"""
|
|
|
|
|
Feature: MindDataset
|
|
|
|
|
Description: Test MindDataset using invalid file (size of mindrecord file header is larger than the upper limit)
|
|
|
|
|
Expectation: Exception is raised as expected
|
|
|
|
|
"""
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
with open(file_name, 'w') as f:
|
2020-03-27 14:49:12 +08:00
|
|
|
f.write('just for test')
|
|
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
2022-10-20 20:51:50 +08:00
|
|
|
with pytest.raises(RuntimeError, match="Invalid file, the size of mindrecord file header "
|
2021-11-30 10:41:05 +08:00
|
|
|
"is larger than the upper limit."):
|
2021-10-19 16:52:57 +08:00
|
|
|
data_set = ds.MindDataset(file_name, columns_list, num_readers)
|
2020-09-05 10:56:38 +08:00
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
2021-08-28 11:03:18 +08:00
|
|
|
pass
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
2020-03-27 14:49:12 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_minddataset_lack_db():
|
2022-06-03 09:24:09 +08:00
|
|
|
"""
|
|
|
|
|
Feature: MindDataset
|
|
|
|
|
Description: Test MindDataset without .db files
|
|
|
|
|
Expectation: Exception is raised as expected
|
|
|
|
|
"""
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2020-03-27 14:49:12 +08:00
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
2022-04-12 16:12:59 +08:00
|
|
|
with pytest.raises(RuntimeError, match=".db exists and do not rename the mindrecord file and meta file."):
|
2021-10-19 16:52:57 +08:00
|
|
|
data_set = ds.MindDataset(file_name, columns_list, num_readers)
|
2020-03-27 14:49:12 +08:00
|
|
|
num_iter = 0
|
2020-09-05 10:56:38 +08:00
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
2020-03-27 14:49:12 +08:00
|
|
|
num_iter += 1
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
2020-04-23 12:02:47 +08:00
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_cv_minddataset_pk_sample_error_class_column():
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
2020-04-23 12:02:47 +08:00
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
2020-09-11 06:08:51 +08:00
|
|
|
sampler = ds.PKSampler(5, None, True, 'no_exist_column')
|
2021-11-30 11:43:17 +08:00
|
|
|
with pytest.raises(RuntimeError, match="Invalid data, 'class_column': no_exist_column can not found "
|
|
|
|
|
"in fields of mindrecord files. Please check 'class_column' in PKSampler"):
|
2021-09-14 20:45:47 +08:00
|
|
|
data_set = ds.MindDataset(
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name, columns_list, num_readers, sampler=sampler)
|
2020-04-23 12:02:47 +08:00
|
|
|
num_iter = 0
|
2020-09-05 10:56:38 +08:00
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
2020-04-23 12:02:47 +08:00
|
|
|
num_iter += 1
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2020-04-23 12:02:47 +08:00
|
|
|
|
2020-05-18 10:31:46 +08:00
|
|
|
|
2020-05-07 14:53:41 +08:00
|
|
|
def test_cv_minddataset_pk_sample_exclusive_shuffle():
|
2022-06-03 09:24:09 +08:00
|
|
|
"""
|
|
|
|
|
Feature: MindDataset
|
|
|
|
|
Description: Test MindDataset by specifying sampler and shuffle at the same time
|
|
|
|
|
Expectation: Exception is raised as expected
|
|
|
|
|
"""
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
2020-05-07 14:53:41 +08:00
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
|
|
|
|
sampler = ds.PKSampler(2)
|
2020-05-14 10:04:51 +08:00
|
|
|
with pytest.raises(Exception, match="sampler and shuffle cannot be specified at the same time."):
|
2021-10-19 16:52:57 +08:00
|
|
|
data_set = ds.MindDataset(file_name, columns_list, num_readers,
|
2020-05-18 10:31:46 +08:00
|
|
|
sampler=sampler, shuffle=False)
|
2020-05-07 14:53:41 +08:00
|
|
|
num_iter = 0
|
2020-09-05 10:56:38 +08:00
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
2020-05-07 14:53:41 +08:00
|
|
|
num_iter += 1
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2020-05-07 14:53:41 +08:00
|
|
|
|
2020-05-18 10:31:46 +08:00
|
|
|
|
2020-05-08 16:11:59 +08:00
|
|
|
def test_cv_minddataset_reader_different_schema():
|
2022-06-03 09:24:09 +08:00
|
|
|
"""
|
|
|
|
|
Feature: MindDataset
|
|
|
|
|
Description: Test MindDataset by including a file that has a different schema from the others
|
|
|
|
|
Expectation: Exception is raised as expected
|
|
|
|
|
"""
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
file_name_1 = file_name + '_1'
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
|
|
|
|
create_diff_schema_cv_mindrecord(file_name_1, 1)
|
2020-05-08 16:11:59 +08:00
|
|
|
columns_list = ["data", "label"]
|
|
|
|
|
num_readers = 4
|
2021-11-30 11:43:17 +08:00
|
|
|
with pytest.raises(RuntimeError, match="Invalid file, the metadata of mindrecord file: "
|
|
|
|
|
"test_cv_minddataset_reader_different_schema_1 is different from others, "
|
|
|
|
|
"please make sure all the mindrecord files generated by the same script."):
|
2021-10-19 16:52:57 +08:00
|
|
|
data_set = ds.MindDataset([file_name, file_name_1], columns_list,
|
2020-05-18 10:31:46 +08:00
|
|
|
num_readers)
|
2020-05-08 16:11:59 +08:00
|
|
|
num_iter = 0
|
2020-08-26 07:52:53 +08:00
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1):
|
2020-05-08 16:11:59 +08:00
|
|
|
num_iter += 1
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
|
|
|
|
os.remove(file_name_1)
|
|
|
|
|
os.remove("{}.db".format(file_name_1))
|
2020-05-08 16:11:59 +08:00
|
|
|
|
2020-05-18 10:31:46 +08:00
|
|
|
|
2020-05-08 16:11:59 +08:00
|
|
|
def test_cv_minddataset_reader_different_page_size():
|
2022-06-03 09:24:09 +08:00
|
|
|
"""
|
|
|
|
|
Feature: MindDataset
|
|
|
|
|
Description: Test MindDataset where one of the files has a different page size
|
|
|
|
|
Expectation: Exception is raised as expected
|
|
|
|
|
"""
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
file_name_1 = file_name + '_1'
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
|
|
|
|
create_diff_page_size_cv_mindrecord(file_name_1, 1)
|
2020-05-08 16:11:59 +08:00
|
|
|
columns_list = ["data", "label"]
|
|
|
|
|
num_readers = 4
|
2021-11-30 11:43:17 +08:00
|
|
|
with pytest.raises(RuntimeError, match="Invalid file, the metadata of mindrecord file: " \
|
|
|
|
|
"test_cv_minddataset_reader_different_page_size_1 is different " \
|
|
|
|
|
"from others, please make sure all " \
|
|
|
|
|
"the mindrecord files generated by the same script."):
|
2021-10-19 16:52:57 +08:00
|
|
|
data_set = ds.MindDataset([file_name, file_name_1], columns_list,
|
2020-05-18 10:31:46 +08:00
|
|
|
num_readers)
|
2020-05-08 16:11:59 +08:00
|
|
|
num_iter = 0
|
2020-08-26 07:52:53 +08:00
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1):
|
2020-05-08 16:11:59 +08:00
|
|
|
num_iter += 1
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
|
|
|
|
os.remove(file_name_1)
|
|
|
|
|
os.remove("{}.db".format(file_name_1))
|
2020-05-14 10:04:51 +08:00
|
|
|
|
2020-05-18 10:31:46 +08:00
|
|
|
|
2020-05-14 10:04:51 +08:00
|
|
|
def test_minddataset_invalidate_num_shards():
|
2022-06-03 09:24:09 +08:00
|
|
|
"""
|
|
|
|
|
Feature: MindDataset
|
|
|
|
|
Description: Test MindDataset where num_shards is invalid
|
|
|
|
|
Expectation: Exception is raised as expected
|
|
|
|
|
"""
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
2020-05-14 10:04:51 +08:00
|
|
|
columns_list = ["data", "label"]
|
|
|
|
|
num_readers = 4
|
2020-06-19 21:58:38 +08:00
|
|
|
with pytest.raises(Exception) as error_info:
|
2021-09-14 20:45:47 +08:00
|
|
|
data_set = ds.MindDataset(
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name, columns_list, num_readers, True, 1, 2)
|
2020-05-14 10:04:51 +08:00
|
|
|
num_iter = 0
|
2020-08-26 07:52:53 +08:00
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1):
|
2020-05-14 10:04:51 +08:00
|
|
|
num_iter += 1
|
2020-07-18 02:06:14 +08:00
|
|
|
try:
|
2021-09-14 20:45:47 +08:00
|
|
|
assert 'Input shard_id is not within the required interval of [0, 0].' in str(
|
|
|
|
|
error_info.value)
|
2020-07-18 02:06:14 +08:00
|
|
|
except Exception as error:
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2020-07-18 02:06:14 +08:00
|
|
|
raise error
|
|
|
|
|
else:
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2020-06-19 21:58:38 +08:00
|
|
|
|
2020-05-14 10:04:51 +08:00
|
|
|
|
|
|
|
|
def test_minddataset_invalidate_shard_id():
|
2022-06-03 09:24:09 +08:00
|
|
|
"""
|
|
|
|
|
Feature: MindDataset
|
|
|
|
|
Description: Test MindDataset where shard_id is invalid
|
|
|
|
|
Expectation: Exception is raised as expected
|
|
|
|
|
"""
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
2020-05-14 10:04:51 +08:00
|
|
|
columns_list = ["data", "label"]
|
|
|
|
|
num_readers = 4
|
2020-06-19 21:58:38 +08:00
|
|
|
with pytest.raises(Exception) as error_info:
|
2021-09-14 20:45:47 +08:00
|
|
|
data_set = ds.MindDataset(
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name, columns_list, num_readers, True, 1, -1)
|
2020-05-14 10:04:51 +08:00
|
|
|
num_iter = 0
|
2020-08-26 07:52:53 +08:00
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1):
|
2020-05-14 10:04:51 +08:00
|
|
|
num_iter += 1
|
2020-07-18 02:06:14 +08:00
|
|
|
try:
|
2021-09-14 20:45:47 +08:00
|
|
|
assert 'Input shard_id is not within the required interval of [0, 0].' in str(
|
|
|
|
|
error_info.value)
|
2020-07-18 02:06:14 +08:00
|
|
|
except Exception as error:
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2020-07-18 02:06:14 +08:00
|
|
|
raise error
|
|
|
|
|
else:
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2020-05-14 10:04:51 +08:00
|
|
|
|
2020-05-18 10:31:46 +08:00
|
|
|
|
2020-05-14 10:04:51 +08:00
|
|
|
def test_minddataset_shard_id_bigger_than_num_shard():
|
2022-06-03 09:24:09 +08:00
|
|
|
"""
|
|
|
|
|
Feature: MindDataset
|
|
|
|
|
Description: Test MindDataset where shard_id is bigger than num_shards
|
|
|
|
|
Expectation: Exception is raised as expected
|
|
|
|
|
"""
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
2020-05-14 10:04:51 +08:00
|
|
|
columns_list = ["data", "label"]
|
|
|
|
|
num_readers = 4
|
2020-06-19 21:58:38 +08:00
|
|
|
with pytest.raises(Exception) as error_info:
|
2021-09-14 20:45:47 +08:00
|
|
|
data_set = ds.MindDataset(
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name, columns_list, num_readers, True, 2, 2)
|
2020-05-14 10:04:51 +08:00
|
|
|
num_iter = 0
|
2020-08-26 07:52:53 +08:00
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1):
|
2020-05-14 10:04:51 +08:00
|
|
|
num_iter += 1
|
2020-07-18 02:06:14 +08:00
|
|
|
try:
|
2021-09-14 20:45:47 +08:00
|
|
|
assert 'Input shard_id is not within the required interval of [0, 1].' in str(
|
|
|
|
|
error_info.value)
|
2020-07-18 02:06:14 +08:00
|
|
|
except Exception as error:
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2020-07-18 02:06:14 +08:00
|
|
|
raise error
|
2020-05-14 10:04:51 +08:00
|
|
|
|
2020-06-19 21:58:38 +08:00
|
|
|
with pytest.raises(Exception) as error_info:
|
2021-09-14 20:45:47 +08:00
|
|
|
data_set = ds.MindDataset(
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name, columns_list, num_readers, True, 2, 5)
|
2020-05-14 10:04:51 +08:00
|
|
|
num_iter = 0
|
2020-08-26 07:52:53 +08:00
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1):
|
2020-05-14 10:04:51 +08:00
|
|
|
num_iter += 1
|
2020-07-18 02:06:14 +08:00
|
|
|
try:
|
2021-09-14 20:45:47 +08:00
|
|
|
assert 'Input shard_id is not within the required interval of [0, 1].' in str(
|
|
|
|
|
error_info.value)
|
2020-07-18 02:06:14 +08:00
|
|
|
except Exception as error:
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2020-07-18 02:06:14 +08:00
|
|
|
raise error
|
|
|
|
|
else:
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2020-05-14 10:04:51 +08:00
|
|
|
|
2020-07-16 19:07:29 +08:00
|
|
|
|
|
|
|
|
def test_cv_minddataset_partition_num_samples_equals_0():
|
2022-06-03 09:24:09 +08:00
|
|
|
"""
|
|
|
|
|
Feature: MindDataset
|
|
|
|
|
Description: Test MindDataset where num_samples is invalid
|
|
|
|
|
Expectation: Exception is raised as expected
|
|
|
|
|
"""
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
2020-07-16 19:07:29 +08:00
|
|
|
columns_list = ["data", "label"]
|
|
|
|
|
num_readers = 4
|
|
|
|
|
|
|
|
|
|
def partitions(num_shards):
|
|
|
|
|
for partition_id in range(num_shards):
|
2021-10-19 16:52:57 +08:00
|
|
|
data_set = ds.MindDataset(file_name, columns_list, num_readers,
|
2020-07-16 19:07:29 +08:00
|
|
|
num_shards=num_shards,
|
2020-12-23 06:56:35 +08:00
|
|
|
shard_id=partition_id, num_samples=-1)
|
2020-07-16 19:07:29 +08:00
|
|
|
num_iter = 0
|
2020-08-26 07:52:53 +08:00
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1):
|
2020-07-16 19:07:29 +08:00
|
|
|
num_iter += 1
|
2021-03-11 11:57:24 +08:00
|
|
|
|
2020-12-23 06:56:35 +08:00
|
|
|
with pytest.raises(ValueError) as error_info:
|
2020-07-16 19:07:29 +08:00
|
|
|
partitions(5)
|
2020-07-18 02:06:14 +08:00
|
|
|
try:
|
2021-09-14 20:45:47 +08:00
|
|
|
assert 'num_samples exceeds the boundary between 0 and 9223372036854775807(INT64_MAX)' in str(
|
|
|
|
|
error_info.value)
|
2020-07-18 02:06:14 +08:00
|
|
|
except Exception as error:
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2020-07-18 02:06:14 +08:00
|
|
|
raise error
|
|
|
|
|
else:
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2020-07-16 19:07:29 +08:00
|
|
|
|
2021-03-11 11:57:24 +08:00
|
|
|
|
2021-01-06 14:09:21 +08:00
|
|
|
def test_mindrecord_exception():
|
2022-06-03 09:24:09 +08:00
|
|
|
"""
|
|
|
|
|
Feature: MindDataset
|
|
|
|
|
Description: Test MindDataset by mapping function that will raise Exception and print error info
|
|
|
|
|
Expectation: Exception is raised as expected
|
|
|
|
|
"""
|
2021-01-06 14:09:21 +08:00
|
|
|
def exception_func(item):
|
|
|
|
|
raise Exception("Error occur!")
|
|
|
|
|
|
2021-10-19 16:52:57 +08:00
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
2021-01-06 14:09:21 +08:00
|
|
|
columns_list = ["data", "file_name", "label"]
|
2022-10-20 20:51:50 +08:00
|
|
|
with pytest.raises(RuntimeError, match="The corresponding data file is"):
|
2021-10-19 16:52:57 +08:00
|
|
|
data_set = ds.MindDataset(file_name, columns_list, shuffle=False)
|
2021-09-14 20:45:47 +08:00
|
|
|
data_set = data_set.map(operations=exception_func, input_columns=["data"],
|
|
|
|
|
num_parallel_workers=1)
|
2021-01-06 14:09:21 +08:00
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
2022-10-20 20:51:50 +08:00
|
|
|
with pytest.raises(RuntimeError, match="The corresponding data file is"):
|
2021-10-19 16:52:57 +08:00
|
|
|
data_set = ds.MindDataset(file_name, columns_list, shuffle=False)
|
2021-09-14 20:45:47 +08:00
|
|
|
data_set = data_set.map(operations=exception_func, input_columns=["file_name"],
|
|
|
|
|
num_parallel_workers=1)
|
2021-01-06 14:09:21 +08:00
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
2022-10-20 20:51:50 +08:00
|
|
|
with pytest.raises(RuntimeError, match="The corresponding data file is"):
|
2021-10-19 16:52:57 +08:00
|
|
|
data_set = ds.MindDataset(file_name, columns_list, shuffle=False)
|
2021-09-14 20:45:47 +08:00
|
|
|
data_set = data_set.map(operations=exception_func, input_columns=["label"],
|
|
|
|
|
num_parallel_workers=1)
|
2021-01-06 14:09:21 +08:00
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
2021-10-19 16:52:57 +08:00
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove("{}.db".format(file_name))
|
2021-01-06 14:09:21 +08:00
|
|
|
|
2021-11-08 14:50:21 +08:00
|
|
|
def test_shuffle_with_num_samples_exception():
|
|
|
|
|
"""
|
2022-05-27 20:53:41 +08:00
|
|
|
Feature: Shuffle files or shuffle samples of each file
|
|
|
|
|
Description: Set Shuffle.FILES or Shuffle.INFILE and num_samples
|
|
|
|
|
Expectation: Exception occurred
|
2021-11-08 14:50:21 +08:00
|
|
|
"""
|
|
|
|
|
MIND_DIR = "../data/mindrecord/testMindDataSet/testImageNetData/imagenet.mindrecord0"
|
|
|
|
|
with pytest.raises(ValueError, match="'Shuffle.FILES' or 'Shuffle.INFILE' and 'num_samples' "
|
|
|
|
|
"cannot be specified at the same time."):
|
|
|
|
|
_ = ds.MindDataset(MIND_DIR, shuffle=ds.Shuffle.FILES, num_samples=5)
|
|
|
|
|
|
|
|
|
|
with pytest.raises(ValueError, match="'Shuffle.FILES' or 'Shuffle.INFILE' and 'num_samples' "
|
|
|
|
|
"cannot be specified at the same time."):
|
|
|
|
|
_ = ds.MindDataset(MIND_DIR, shuffle=ds.Shuffle.INFILE, num_samples=5)
|
2021-01-06 14:09:21 +08:00
|
|
|
|
2022-04-12 16:12:59 +08:00
|
|
|
|
|
|
|
|
def test_rename_exception_01():
|
|
|
|
|
"""
|
2022-05-27 20:53:41 +08:00
|
|
|
Feature: Rename mindrecord file
|
|
|
|
|
Description: Dataset that contains single mindrecord file
|
|
|
|
|
Expectation: Exception occurred
|
2022-04-12 16:12:59 +08:00
|
|
|
"""
|
|
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
|
|
|
|
|
|
|
|
|
new_file_name = file_name + "_new"
|
|
|
|
|
|
|
|
|
|
os.rename(file_name, new_file_name)
|
|
|
|
|
|
|
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
|
|
|
|
with pytest.raises(RuntimeError, match="can not be found. Please check whether the mindrecord file exists" \
|
|
|
|
|
" and do not rename the mindrecord file."):
|
|
|
|
|
data_set = ds.MindDataset(new_file_name, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
with pytest.raises(RuntimeError, match=".db exists and do not rename the mindrecord file and meta file."):
|
|
|
|
|
data_set = ds.MindDataset([new_file_name], columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
os.remove(new_file_name)
|
|
|
|
|
os.remove(file_name + ".db")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_rename_exception_02():
|
|
|
|
|
"""
|
2022-05-27 20:53:41 +08:00
|
|
|
Feature: Rename mindrecord meta file
|
|
|
|
|
Description: Dataset that contains single mindrecord file
|
|
|
|
|
Expectation: Exception occurred
|
2022-04-12 16:12:59 +08:00
|
|
|
"""
|
|
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
|
|
|
|
|
|
|
|
|
new_file_name = file_name + "_new"
|
|
|
|
|
os.rename(file_name + ".db", new_file_name + ".db")
|
|
|
|
|
|
|
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
|
|
|
|
with pytest.raises(RuntimeError, match=".db exists and do not rename the mindrecord file and meta file."):
|
|
|
|
|
data_set = ds.MindDataset(file_name, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
with pytest.raises(RuntimeError, match=".db exists and do not rename the mindrecord file and meta file."):
|
|
|
|
|
data_set = ds.MindDataset([file_name], columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
os.remove(file_name)
|
|
|
|
|
os.remove(new_file_name + ".db")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_rename_exception_03():
|
|
|
|
|
"""
|
2022-05-27 20:53:41 +08:00
|
|
|
Feature: Rename both mindrecord file and meta file
|
|
|
|
|
Description: Dataset that contains single mindrecord file
|
|
|
|
|
Expectation: Exception occurred
|
2022-04-12 16:12:59 +08:00
|
|
|
"""
|
|
|
|
|
file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(file_name, 1)
|
|
|
|
|
|
|
|
|
|
new_file_name = file_name + "_new"
|
|
|
|
|
|
|
|
|
|
os.rename(file_name, new_file_name)
|
|
|
|
|
os.rename(file_name + ".db", new_file_name + ".db")
|
|
|
|
|
|
|
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
|
|
|
|
with pytest.raises(RuntimeError, match="can not be found. Please check whether the mindrecord file exists" \
|
|
|
|
|
" and do not rename the mindrecord file."):
|
|
|
|
|
data_set = ds.MindDataset(new_file_name, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
with pytest.raises(RuntimeError, match="can not match. Please do not rename the mindrecord file or meta file."):
|
|
|
|
|
data_set = ds.MindDataset([new_file_name], columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
os.remove(new_file_name)
|
|
|
|
|
os.remove(new_file_name + ".db")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_rename_exception_04():
|
|
|
|
|
"""
|
2022-05-27 20:53:41 +08:00
|
|
|
Feature: Rename current mindrecord file
|
|
|
|
|
Description: Dataset that contains multiple mindrecord files
|
|
|
|
|
Expectation: Exception occurred
|
2022-04-12 16:12:59 +08:00
|
|
|
"""
|
|
|
|
|
ori_file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(ori_file_name, 4)
|
|
|
|
|
|
|
|
|
|
file_name = ori_file_name + '0'
|
|
|
|
|
new_file_name = file_name + "_new"
|
|
|
|
|
|
|
|
|
|
os.rename(file_name, new_file_name)
|
|
|
|
|
|
|
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
|
|
|
|
with pytest.raises(RuntimeError, match="can not be found. Please check whether the mindrecord file exists" \
|
|
|
|
|
" and do not rename the mindrecord file."):
|
|
|
|
|
data_set = ds.MindDataset(new_file_name, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
file_list = [ori_file_name + str(x) for x in range(4)]
|
|
|
|
|
file_list[0] = new_file_name
|
|
|
|
|
with pytest.raises(RuntimeError, match=".db exists and do not rename the mindrecord file and meta file."):
|
|
|
|
|
data_set = ds.MindDataset(file_list, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
os.remove(new_file_name)
|
|
|
|
|
for x in range(4):
|
|
|
|
|
if os.path.exists(ori_file_name + str(x)):
|
|
|
|
|
os.remove(ori_file_name + str(x))
|
|
|
|
|
if os.path.exists(ori_file_name + str(x) + ".db"):
|
|
|
|
|
os.remove(ori_file_name + str(x) + ".db")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_rename_exception_05():
|
|
|
|
|
"""
|
2022-05-27 20:53:41 +08:00
|
|
|
Feature: Rename other mindrecord file
|
|
|
|
|
Description: Dataset that contains multiple mindrecord files
|
|
|
|
|
Expectation: Exception occurred
|
2022-04-12 16:12:59 +08:00
|
|
|
"""
|
|
|
|
|
ori_file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(ori_file_name, 4)
|
|
|
|
|
|
|
|
|
|
other_file_name = ori_file_name + '2'
|
|
|
|
|
new_file_name = other_file_name + "_new"
|
|
|
|
|
|
|
|
|
|
os.rename(other_file_name, new_file_name)
|
|
|
|
|
|
|
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
|
|
|
|
file_name = ori_file_name + '0'
|
|
|
|
|
with pytest.raises(RuntimeError, match="can not be found. Please check whether the mindrecord file exists" \
|
|
|
|
|
" and do not rename the mindrecord file."):
|
|
|
|
|
data_set = ds.MindDataset(file_name, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
file_list = [ori_file_name + str(x) for x in range(4)]
|
|
|
|
|
file_list[2] = new_file_name
|
|
|
|
|
with pytest.raises(RuntimeError, match=".db exists and do not rename the mindrecord file and meta file."):
|
|
|
|
|
data_set = ds.MindDataset(file_list, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
os.remove(new_file_name)
|
|
|
|
|
for x in range(4):
|
|
|
|
|
if os.path.exists(ori_file_name + str(x)):
|
|
|
|
|
os.remove(ori_file_name + str(x))
|
|
|
|
|
if os.path.exists(ori_file_name + str(x) + ".db"):
|
|
|
|
|
os.remove(ori_file_name + str(x) + ".db")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_rename_exception_06():
|
|
|
|
|
"""
|
2022-05-27 20:53:41 +08:00
|
|
|
Feature: Rename current meta file
|
|
|
|
|
Description: Dataset that contains multiple mindrecord files
|
|
|
|
|
Expectation: Exception occurred
|
2022-04-12 16:12:59 +08:00
|
|
|
"""
|
|
|
|
|
ori_file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(ori_file_name, 4)
|
|
|
|
|
|
|
|
|
|
file_name = ori_file_name + '0'
|
|
|
|
|
new_file_name = file_name + "_new"
|
|
|
|
|
|
|
|
|
|
os.rename(file_name + ".db", new_file_name + ".db")
|
|
|
|
|
|
|
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
|
|
|
|
with pytest.raises(RuntimeError, match=".db exists and do not rename the mindrecord file and meta file."):
|
|
|
|
|
data_set = ds.MindDataset(file_name, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
file_list = [ori_file_name + str(x) for x in range(4)]
|
|
|
|
|
with pytest.raises(RuntimeError, match=".db exists and do not rename the mindrecord file and meta file."):
|
|
|
|
|
data_set = ds.MindDataset(file_list, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
os.remove(new_file_name + ".db")
|
|
|
|
|
for x in range(4):
|
|
|
|
|
if os.path.exists(ori_file_name + str(x)):
|
|
|
|
|
os.remove(ori_file_name + str(x))
|
|
|
|
|
if os.path.exists(ori_file_name + str(x) + ".db"):
|
|
|
|
|
os.remove(ori_file_name + str(x) + ".db")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_rename_exception_07():
|
|
|
|
|
"""
|
2022-05-27 20:53:41 +08:00
|
|
|
Feature: Rename other meta file
|
|
|
|
|
Description: Dataset that contains multiple mindrecord files
|
|
|
|
|
Expectation: Exception occurred
|
2022-04-12 16:12:59 +08:00
|
|
|
"""
|
|
|
|
|
ori_file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(ori_file_name, 4)
|
|
|
|
|
|
|
|
|
|
other_file_name = ori_file_name + '2'
|
|
|
|
|
new_file_name = other_file_name + "_new"
|
|
|
|
|
|
|
|
|
|
os.rename(other_file_name + ".db", new_file_name + ".db")
|
|
|
|
|
|
|
|
|
|
file_name = ori_file_name + '0'
|
|
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
|
|
|
|
with pytest.raises(RuntimeError, match=".db exists and do not rename the mindrecord file and meta file."):
|
|
|
|
|
data_set = ds.MindDataset(file_name, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
file_list = [ori_file_name + str(x) for x in range(4)]
|
|
|
|
|
with pytest.raises(RuntimeError, match=".db exists and do not rename the mindrecord file and meta file."):
|
|
|
|
|
data_set = ds.MindDataset(file_list, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
os.remove(new_file_name + ".db")
|
|
|
|
|
for x in range(4):
|
|
|
|
|
if os.path.exists(ori_file_name + str(x)):
|
|
|
|
|
os.remove(ori_file_name + str(x))
|
|
|
|
|
if os.path.exists(ori_file_name + str(x) + ".db"):
|
|
|
|
|
os.remove(ori_file_name + str(x) + ".db")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_rename_exception_08():
|
|
|
|
|
"""
|
2022-05-27 20:53:41 +08:00
|
|
|
Feature: Rename both current mindrecord file and meta file
|
|
|
|
|
Description: Dataset that contains multiple mindrecord files
|
|
|
|
|
Expectation: Exception occurred
|
2022-04-12 16:12:59 +08:00
|
|
|
"""
|
|
|
|
|
ori_file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(ori_file_name, 4)
|
|
|
|
|
|
|
|
|
|
file_name = ori_file_name + '0'
|
|
|
|
|
new_file_name = file_name + "_new"
|
|
|
|
|
|
|
|
|
|
os.rename(file_name, new_file_name)
|
|
|
|
|
os.rename(file_name + ".db", new_file_name + ".db")
|
|
|
|
|
|
|
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
|
|
|
|
with pytest.raises(RuntimeError, match="can not be found. Please check whether the mindrecord file exists" \
|
|
|
|
|
" and do not rename the mindrecord file."):
|
|
|
|
|
data_set = ds.MindDataset(new_file_name, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
file_list = [ori_file_name + str(x) for x in range(4)]
|
|
|
|
|
file_list[0] = new_file_name
|
|
|
|
|
with pytest.raises(RuntimeError, match="can not match. Please do not rename the mindrecord file or meta file."):
|
|
|
|
|
data_set = ds.MindDataset(file_list, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
os.remove(new_file_name)
|
|
|
|
|
os.remove(new_file_name + ".db")
|
|
|
|
|
for x in range(4):
|
|
|
|
|
if os.path.exists(ori_file_name + str(x)):
|
|
|
|
|
os.remove(ori_file_name + str(x))
|
|
|
|
|
if os.path.exists(ori_file_name + str(x) + ".db"):
|
|
|
|
|
os.remove(ori_file_name + str(x) + ".db")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
def test_rename_exception_09():
|
|
|
|
|
"""
|
2022-05-27 20:53:41 +08:00
|
|
|
Feature: Rename both other mindrecord file and meta file
|
|
|
|
|
Description: Dataset that contains multiple mindrecord files
|
|
|
|
|
Expectation: Exception occurred
|
2022-04-12 16:12:59 +08:00
|
|
|
"""
|
|
|
|
|
ori_file_name = os.environ.get('PYTEST_CURRENT_TEST').split(':')[-1].split(' ')[0]
|
|
|
|
|
create_cv_mindrecord(ori_file_name, 4)
|
|
|
|
|
|
|
|
|
|
other_file_name = ori_file_name + '2'
|
|
|
|
|
new_file_name = other_file_name + "_new"
|
|
|
|
|
|
|
|
|
|
os.rename(other_file_name, new_file_name)
|
|
|
|
|
os.rename(other_file_name + ".db", new_file_name + ".db")
|
|
|
|
|
|
|
|
|
|
columns_list = ["data", "file_name", "label"]
|
|
|
|
|
num_readers = 4
|
|
|
|
|
file_name = ori_file_name + '0'
|
|
|
|
|
with pytest.raises(RuntimeError, match="can not be found. Please check whether the mindrecord file exists" \
|
|
|
|
|
" and do not rename the mindrecord file."):
|
|
|
|
|
data_set = ds.MindDataset(file_name, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
file_list = [ori_file_name + str(x) for x in range(4)]
|
|
|
|
|
file_list[2] = new_file_name
|
|
|
|
|
with pytest.raises(RuntimeError, match="can not match. Please do not rename the mindrecord file or meta file."):
|
|
|
|
|
data_set = ds.MindDataset(file_list, columns_list, num_readers)
|
|
|
|
|
num_iter = 0
|
|
|
|
|
for _ in data_set.create_dict_iterator(num_epochs=1, output_numpy=True):
|
|
|
|
|
num_iter += 1
|
|
|
|
|
|
|
|
|
|
os.remove(new_file_name)
|
|
|
|
|
os.remove(new_file_name + ".db")
|
|
|
|
|
for x in range(4):
|
|
|
|
|
if os.path.exists(ori_file_name + str(x)):
|
|
|
|
|
os.remove(ori_file_name + str(x))
|
|
|
|
|
if os.path.exists(ori_file_name + str(x) + ".db"):
|
|
|
|
|
os.remove(ori_file_name + str(x) + ".db")
|
|
|
|
|
|
|
|
|
|
|
2020-07-18 02:06:14 +08:00
|
|
|
if __name__ == '__main__':
|
|
|
|
|
test_cv_lack_json()
|
|
|
|
|
test_cv_lack_mindrecord()
|
|
|
|
|
test_invalid_mindrecord()
|
|
|
|
|
test_minddataset_lack_db()
|
|
|
|
|
test_cv_minddataset_pk_sample_error_class_column()
|
|
|
|
|
test_cv_minddataset_pk_sample_exclusive_shuffle()
|
|
|
|
|
test_cv_minddataset_reader_different_schema()
|
|
|
|
|
test_cv_minddataset_reader_different_page_size()
|
|
|
|
|
test_minddataset_invalidate_num_shards()
|
|
|
|
|
test_minddataset_invalidate_shard_id()
|
|
|
|
|
test_minddataset_shard_id_bigger_than_num_shard()
|
|
|
|
|
test_cv_minddataset_partition_num_samples_equals_0()
|
2021-01-06 14:09:21 +08:00
|
|
|
test_mindrecord_exception()
|
2022-04-12 16:12:59 +08:00
|
|
|
test_rename_exception_01()
|
|
|
|
|
test_rename_exception_02()
|
|
|
|
|
test_rename_exception_03()
|
|
|
|
|
test_rename_exception_04()
|
|
|
|
|
test_rename_exception_05()
|
|
|
|
|
test_rename_exception_06()
|
|
|
|
|
test_rename_exception_07()
|
|
|
|
|
test_rename_exception_08()
|
|
|
|
|
test_rename_exception_09()
|