hummingbird
/
mindarmour
Not watched
2
0
Fork 0
Code
Releases 17 Wiki Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain
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.
601 Commits
18 Branches
193 MB
132 Download
Branch: master
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from 'master'
${ noResults }
mindarmour/requirements.txt

requirements.txt 184 B
Raw Permalink Normal View History

initial version
5 years ago
Update the version of scipy
4 years ago
before rebase, refactor mechanisms implementation. rename mech._decay_policy to mech._noise_update. after rebase, rename norm_clip to norm_bound. change mechanismsfactory to noisemechanismsfactory in test. 11 12 13 1414141414141414141414141414 15
4 years ago
update pillow version
3 years ago
initial version
5 years ago
​
​
update scipy version and scikit-learn version.
4 years ago
add easydict to requirements.txt
4 years ago
update pillow version
3 years ago
 12345678910 
  1. numpy >= 1.17.0

  2. scipy >= 1.5.2

  3. matplotlib >= 3.2.1

  4. pillow >= 7.2.0

  5. pytest >= 4.3.1

  6. wheel >= 0.32.0

  7. setuptools >= 40.8.0

  8. scikit-learn >= 0.23.1

  9. easydict >= 1.9

  10. opencv-python >= 4.1.2.30

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。