Plotting TF-IDF and K-Means Data with Matplotlib (Topic Modeling in Python for DH 02.04)

[@luca-o-sharp]

This is such a beauty! Thanks for sharing it all in such an easy to understand way!

[@python-programming]

Thanks! No problem at all! So happy others are finding it useful!

[@felixgiles2617]

Great video, thank you! In case of overlapping data points I think you would get better results too if you call model.fit_predict() with scatter_plot_points from line 112 (reduced vectors) for the kmean_indices. It also helps for plotting the centroids of the clusters.

[@python-programming]

Thanks for mentioning this! I will check this out!

[@MrSerazzo]

You saved my life

[@rafaanimir8820]

Thank you so much for this series I have learned so much from you.

[@python-programming]

I am so happy to hear that!! Thanks!

[@miketueller5916]

You mention in the video that the description will have a link to stackoverflow. I don't see it right now. Can you add that? Thanks!

[@python-programming]

Thanks for catching that! Here you are: stackoverflow.com/questions/27494202/how-do-i-visualize-data-points-of-tf-idf-vectors-for-kmeans-clustering -- I'm adding it to the description now.

[@maxrichardson5252]

hello, every time i remake the graph the colours change for the plot why does it do this and how would i fix

[@KatorZChannel]

What i need to do with a large data?

[@fancynex]

I have an IndexError: list index out of range =(

[@irreadings]

same here. colors[d] doesn't work because d sometimes has a value larger than the length of the colors list.

[@irreadings]

Oh, I got what the problem was. He changed the "true_k" variable to 5 (whereas formerly it was 20).

[@fancynex]

​@@irreadings thank you! I'll even go back to the code and fix it so that it finally works =)