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.

config.py 1.4 kB

123456789101112131415161718192021222324252627282930313233343536373839404142434445
  1. # Copyright 2020 Huawei Technologies Co., Ltd
  2. #
  3. # Licensed under the Apache License, Version 2.0 (the "License");
  4. # you may not use this file except in compliance with the License.
  5. # You may obtain a copy of the License at
  6. #
  7. # http://www.apache.org/licenses/LICENSE-2.0
  8. #
  9. # Unless required by applicable law or agreed to in writing, software
  10. # distributed under the License is distributed on an "AS IS" BASIS,
  11. # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  12. # See the License for the specific language governing permissions and
  13. # limitations under the License.
  14. # ============================================================================
  15. """
  16. network config setting, will be used in train.py and eval.py
  17. """
  18. from easydict import EasyDict as edict
  19. # config for vgg16, cifar100
  20. cifar_cfg = edict({
  21. "num_classes": 100,
  22. "lr": 0.01,
  23. "lr_init": 0.01,
  24. "lr_max": 0.1,
  25. "lr_epochs": '30,60,90,120',
  26. "lr_scheduler": "step",
  27. "warmup_epochs": 5,
  28. "batch_size": 64,
  29. "max_epoch": 100,
  30. "momentum": 0.9,
  31. "weight_decay": 5e-4,
  32. "loss_scale": 1.0,
  33. "label_smooth": 0,
  34. "label_smooth_factor": 0,
  35. "buffer_size": 10,
  36. "image_size": '224,224',
  37. "pad_mode": 'same',
  38. "padding": 0,
  39. "has_bias": False,
  40. "batch_norm": True,
  41. "keep_checkpoint_max": 10,
  42. "initialize_mode": "XavierUniform",
  43. "has_dropout": False
  44. })

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。