Yasuhiro Kubota

Hi, there. I faced the similar problem and have solved it. The cause of the problem I faced was that some header's level were NOT increasing consecutively as follows. ```markdown...

Hi, @omerbrandis . It seems the cause of the error is `OriginalModel`. Just in case, I will examine the cause of the problem. Could you please submit the code snippet...

Hi, @roma-glushko . Thank you for sharing your great ideas. > This would dramatically reduce learning curve and increase adoption of the library. I think so too. I believe they...

@OysterSandwich , Thank you for your reporting the error! You can avoid this error by modifying `input_modifiers` and `regularizers` options like below: ```python from tf_keras_vis.utils.regularizers import Norm activations = activation_maximization(score,...

Thank you for letting me know that. > Cannot say I already fully understand your solution, but I can report the former one with jitter is working: > However, the...

> However, the edited one with tf_keras_vis.utils.regularizers import Norm does not work..? I looked into this error, but I didn't find the cause of it. So, if you could, please...

I've leaned it in [keras example](https://keras.io/examples/vision/visualizing_what_convnets_learn/) , [keras-vis](https://github.com/raghakot/keras-vis) , and [the great helpful article to understand Feature Visualization](https://distill.pub/2017/feature-visualization/) . The paper you're looking for may be quoted by above site....

Hi, @omerbrandis . I'm sorry for the late reply. In default, ActivationMaximization is tuned to visualize well with VGG16. So, when you visualize the dense layer of other than VGG16,...

@amizeranschi , Thank you for your request! I believe, as you said, we should publish `tf-keras-vis` to both pip and conda repositories for all users. I will do it in...

I failed, so reopen this issue and am going to investigate the cause.