You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.

package.sh 1.0 kB

5 years ago
1234567891011121314151617181920212223242526272829303132333435363738
  1. #!/bin/bash
  2. # Copyright 2019 Huawei Technologies Co., Ltd
  3. #
  4. # Licensed under the Apache License, Version 2.0 (the "License");
  5. # you may not use this file except in compliance with the License.
  6. # You may obtain a copy of the License at
  7. #
  8. # http://www.apache.org/licenses/LICENSE-2.0
  9. #
  10. # Unless required by applicable law or agreed to in writing, software
  11. # distributed under the License is distributed on an "AS IS" BASIS,
  12. # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  13. # See the License for the specific language governing permissions and
  14. # limitations under the License.
  15. set -e
  16. BASEPATH=$(cd "$(dirname $0)"; pwd)
  17. OUTPUT_PATH="${BASEPATH}/output"
  18. PYTHON=$(which python3)
  19. mk_new_dir() {
  20. local create_dir="$1" # the target to make
  21. if [[ -d "${create_dir}" ]];then
  22. rm -rf "${create_dir}"
  23. fi
  24. mkdir -pv "${create_dir}"
  25. }
  26. mk_new_dir "${OUTPUT_PATH}"
  27. ${PYTHON} ${BASEPATH}/setup.py bdist_wheel
  28. mv ${BASEPATH}/dist/*whl ${OUTPUT_PATH}
  29. echo "------Successfully created mindarmour package------"

MindArmour关注AI的安全和隐私问题。致力于增强模型的安全可信、保护用户的数据隐私。主要包含3个模块:对抗样本鲁棒性模块、Fuzz Testing模块、隐私保护与评估模块。 对抗样本鲁棒性模块 对抗样本鲁棒性模块用于评估模型对于对抗样本的鲁棒性,并提供模型增强方法用于增强模型抗对抗样本攻击的能力,提升模型鲁棒性。对抗样本鲁棒性模块包含了4个子模块:对抗样本的生成、对抗样本的检测、模型防御、攻防评估。