Aurélien Geron

Interesting, thanks for your feedback. ~~I forgot to add Jupyter to `requirements.txt`. I'll do that now, thanks again!~~ **Edit**: mmh, that's not it, Jupyter was there. Could you please tell...

In the thread you pointed to, @Cadair says: > You will have to have `environment_kernels` installed in the Python env (conda, virtualenv or system) where you are running the `jupyter...

Thanks @robmarkcole, could you please share what jupyter-related packages you have? Perhaps more simply, your list of python packages (and their versions).

I really don't see where the problem comes from, and I cannot reproduce it. I believe this is either an Anaconda issue, or a Jupyter issue (or both). Please check...

For example [this open Jupyter issue](https://github.com/jupyter/notebook/issues/2875) seems related.

Hi @odysseus0 , Thanks for your feedback. What they meant by "_no more globals_" is not that you can't use global variables in your own code (even though, I agree...

You're most welcome, @odysseus0. :) Yes, you are exactly right, the global that gets removed is the global scope of `tf.Variable` within the default graph. Whenever you created a `tf.Variable`...

Thanks @darthdeus . The Keras API specifies that there should be a `keras.utils.model_to_dot()` function, but it's only available in `tensorflow.python.keras.utils.vis_utils`. I submitted a fix, hopefully it will be merged soon....

Here's the error output: ``` $ roc test bug.roc An internal compiler expectation was broken. This is definitely a compiler bug. Please file an issue here: Missing module params should've...

For production code, you're totally right, but to avoid increasing the size of the book, and to keep things focused on the main points, I try to focus only on...