Blog Storm

[sambellsoup]

Let's think of things to write about and maybe some of the ideas are good.

[sambellsoup]

How Best to Git

[sambellsoup]

Make your own baby database. From Scraping to SQL. (Scraping was ruled not illegal in Sep 2019)

[sambellsoup]

Basic ML project with music (most likely Midi files, so find database of Midi files). Use ML to create unique midi files which could then be transcribed to sheet music. A part 2 or even 3 could be to use NLP to generate music lyrics. Use ML to create a song for a given artist. Feed the computer lots of Alanis Morissette / Beatles / Billy Joel / other prolific artist songs and then see what it produces. It may be necessary to dissect songs as: melody, bass, harmony, and lyrics.

Songs may be weighted based on record sales (popularity). How to decide on key, time signature, and tempo?

[sambellsoup]

Inventory Allocation. Given a sales history of different products at different locations, how would one go about predicting the allocation of inventory for future orders. How could we accommodate for potential marketing sales or unexpected trends? Clue: standard deviation / area under the curve. What would be the ideal data set? Part 2: How to execute your ideal dataset

[lorarjohns]

General features

• Can make google colab notebooks or similar for tutorials
• Could make a VSCode plugin or use existing ones for pair programming, code challenges, etc.
• Either paid or not, can provide tutorials/online courses to extend coursework on topics in-depth that might be too deep for the general audience but might be valuable for people anyway!

Post categories

Math

• Math for human flourishing: puzzles, algorithms, and math art
• Machine learning math: breaking down core algorithms to make them understandable
• NLP math / translating notation to code
• Statistics and probability

Programming

• Object oriented programming
• Beyond Python: Other languages and why they're cool (I <3 Cython and C and C++ and Rust and Go and I don't <3 JavaScript as much but I use it! Also assembly is a fun thing to talk about)
• I liked your "SQL is Hufflepuff" analogy and that is such a blog post and #mood
• but also, there's a book based on the Art of War that takes the view that SQL is war, and that is also a #mood

DevOps

• Docker
• Kubernetes
• Write your own Python package
• Structuring a python project
• virtual environments (I stg, most important thing)
• GitHub and git
• CI/CD

[lorarjohns]

Basic ML project with music (most likely Midi files, so find database of Midi files). Use ML to create unique midi files which could then be transcribed to sheet music. A part 2 or even 3 could be to use NLP to generate music lyrics. Use ML to create a song for a given artist. Feed the computer lots of Alanis Morissette / Beatles / Billy Joel / other prolific artist songs and then see what it produces. It may be necessary to dissect songs as: melody, bass, harmony, and lyrics.

Songs may be weighted based on record sales (popularity). How to decide on key, time signature, and tempo?

Feed the computer lots of Alanis Morissette / Beatles / Billy Joel / other prolific artist songs and then see what it produces.

I love the idea of you writing Alanis Morissette - Billy Joel mashups with a bot.