In the PCA section there is the following quote:

We see that these 150 components account for just over 90% of the variance.

While not inaccurate (150 componen

Alexnet implementation in tensorflow has incomplete architecture where 2 convolution neural layers are missing. This issue is in reference to the python notebook mentioned below.

https://github.com/donnemartin/data-science-ipython-notebooks/blob/master/deep-learning/tensor-flow-examples/notebooks/3_neural_networks/alexnet.ipynb

As you can see in this example from the documentation docs, there is a linestyles argument, but it doesn't show in the legend. This kind of defies the purpose, printing black/white you still can't tell which is which (except from the markers, but often the markers are hard to read because the line is too wide relativel

On this page at ClassificationReport

For binary classifier, the part “not to label an instance positive that is actually negative” sounded like “to correctly classif

Most of the people who start out new don't find a latest feed of community hyped resources on ML and DL topics. It would be pretty good if we add a Reddit section.
If you're fine with this suggestion I'll put up a PR with the update 😄

Salut,

I'm thinking you should maybe remove the (very) few references to pylab in your tutorial, as the offcial documentation now says "The pylab API (disapproved)"

This code:

https://github.com/nschloe/tikzplotlib/blob/aee7a4ff145159476a7d07248f4b93c2372e73e7/tikzplotlib/axes.py#L207-L209

may produce the axes key 'minor xticklabels' which is not know in PGFPLOTS as of version 1.16

The same holds for y

I got lost again when trying to navigate the cartopy website!

This may come from the fact that some pages are listed on the home page, in the Getting Started section, but some other useful pages are not listed there, and you can only find them by searching the web site, or by visiting one of the pages listed in Getting Started and then clicking on the Next/Previous Topic in the

Example taken from the documentation, adding ha='center' to plt.text() calls:

np.random.seed(0)
x, y = np.random.random((2,30))
fig, ax = plt.subplots()
plt.plot(x, y, 'bo')
texts = [plt.text(x[i], y[i], 'Text%s' %i, ha=

It seems the title is not being animated as the rest of the figure.
In my usage, it just displayed the last value during all video.

Would it make sense to implement a get_cmap function as colorcet.cm.get_cmap that could override Matplotlib's so that external libraries using MPL's get_cmap could just replace it with Colorcet's get_cmap if Colorcet is available at runtime? This would make interfaceing with Colorcet super easy for other libraries already leveraging MPL's colormaps.

I'm thinking that this function would