Discussions

• 1 You must be logged in to vote

  🙏

  Facing issues (loss.backwards not computable if model NOT in training mode) while generating adversarial examples

  SubrangshuDas asked Feb 25, 2024 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  🙏

  CleverHans for ML models with non-neural network architecture

  mahrukhS asked Jan 2, 2024 in Q&A · Unanswered

  

  

  0
• 1 You must be logged in to vote

  💡

  The compute_gradient function has an unnecessary requirement that enforces the presence of the "labels" and "logits" attributes in the provided loss function (loss_fn).

  HaimFisher started Jun 28, 2023 in Ideas

  

  

  0
• 1 You must be logged in to vote

  💬

  4.0.0 carlini_wagner_l2 bug Dimensions must be equal, but are 3 and 20 for '{{node mul_5}} = Mul[T=DT_FLOAT](const, Maximum)' with input shapes: [20,32,32,3], [20].

  inspirepassion started May 28, 2023 in General

  

  

  0
• 2 You must be logged in to vote

  💡

  Guides, tutorials & documentation in Jupyter/Colab Notebooks for v4.0

  axantillon started Mar 1, 2021 in Ideas

  

  

  1