![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Spam, spam, spam...
Book Review: Learning Python, by Mark Lutz
I began reading this book in July. Then, when I got to around page 320, the Kindle edition went blank. I logged a support call, was refunded and told to try again, which I did, with the same result. I discovered the whole text was available on my PC Kindle application, however. Then, some months later, I deleted it from my device again and downloaded it fresh, and this time the whole book appeared. By this point I was already wearing thin of the author's style, which I found verbose, repetitive, and sometimes going to extreme lengths discussing what could only be beginner problems, presumably encountered when running training courses, that would never have occurred to me.
My interest in Python arises because it seems to be prominent in machine learning; I thought I should find out at least the basics on this subject, and an O'Reilly book seemed a logical way to learn the basics about the language itself. Although I hope I have achieved this, unfortunately it's been a very tedious process, as the book doesn't chart a particularly clear course, forever noting differences between Python 2 (which is now more or less officially dead) and 3 and going off on tangents. I come away with mixed impressions of the Python language itself: some aspects strongly hint at a mathematical background, which perhaps explains its use in ML; but others suggest a really hellish mashup, with myriad backwards-incompatible changes, bolted-on object-oriented features, and ever more esoteric interceptions and workarounds. I feel I could pick up a book on machine learning that had Python examples and follow it, but I couldn't really recommend this book as a route to getting there.
I began reading this book in July. Then, when I got to around page 320, the Kindle edition went blank. I logged a support call, was refunded and told to try again, which I did, with the same result. I discovered the whole text was available on my PC Kindle application, however. Then, some months later, I deleted it from my device again and downloaded it fresh, and this time the whole book appeared. By this point I was already wearing thin of the author's style, which I found verbose, repetitive, and sometimes going to extreme lengths discussing what could only be beginner problems, presumably encountered when running training courses, that would never have occurred to me.
My interest in Python arises because it seems to be prominent in machine learning; I thought I should find out at least the basics on this subject, and an O'Reilly book seemed a logical way to learn the basics about the language itself. Although I hope I have achieved this, unfortunately it's been a very tedious process, as the book doesn't chart a particularly clear course, forever noting differences between Python 2 (which is now more or less officially dead) and 3 and going off on tangents. I come away with mixed impressions of the Python language itself: some aspects strongly hint at a mathematical background, which perhaps explains its use in ML; but others suggest a really hellish mashup, with myriad backwards-incompatible changes, bolted-on object-oriented features, and ever more esoteric interceptions and workarounds. I feel I could pick up a book on machine learning that had Python examples and follow it, but I couldn't really recommend this book as a route to getting there.