I have been reading Géron’s Hands-On Machine Learning with Scikit-Learn & TensorFlow. The title says it all. It is a good review on ML topics but above all it is very practical and shows with real examples how to implement ML solutions, from the traditional to the deep learning sort. It is rather thin on theory but that is fine as we have seen other books that tackle that very well already but leave practicalities outside.
I have read a few books on brain and neurology and I was not sure about getting this one by Dutch physician and neurobiologist Dick Swaab, “Wij zijn ons brein” (We are our brain in English). I am glad I did, even though I have some criticism to the book.
There are 22 chapters (including the conclusions) covering topics from brain development, the brain in the womb, sexual differentiation, sexual development, hormones, emotions, consciousness, electrical stimulation, morals, memory, free will, death and evolution. There are some tough stories, a few jokes, lots of brief accounts on research past and present and some general recurring themes that Swaab wants to address.
One thing that shocked me was the author’s profound aversion to sports…unless those sports are chess or bridge. Although he often stresses the importance of evidence his arguments against are extremely biased in this area: he basically refers to physical damage produced by high impact sports. I am not a health specialist but a simple search on the Internet for peer-reviewed articles on the benefits of (moderate) sports do give some interesting results.
I found it was a pity Dr Swaab did not include a biography. I know this is a book for laymen like me but even people like me want once in a while to read a little bit more about this or that.
Dos libros muy diferentes han ocupado algunas horas de mi tiempo libre en las últimas semanas. Uno es un libro de ciencia popular en español, El Ojo Desnudo. En esta publicación Antonio Martínez Ron nos da un buen paseo por un sinnúmero de investigaciones realizadas sobre el ojo y la visión. Hay de todo un poco. Es uno de esos libros de ciencia general que lo pone a pensar a uno en muchas cosas distintas que se nos atisban de repente muy interrelacionadas.
El otro libro, del lado puramente técnico, es Fluent in Python. Está concebido para aquellos que ya dominan este idioma de programación y quieren adentrarse más en tal o cual punto. Está muy bien escrito.
I read Lantz’s Machine Learning with R. It is a rather decent introduction to what it says. The first chapters are an introduction on R and a tiny one on ML. Then it goes into several classification algorithms. There are some chapters on clustering, decision trees, the usual stuff. I got most profit from the chapter on neural networks and the final topics on model performance. I wish the book had gone deeper into some R packages, but I suppose the author wanted to keep it open for those who were just getting into ML.
Now I am delving into Goodfellow’s Deep Learning and I am enjoying it. First I was just trying the on-line version, but I got now the paper one: this is a good reference, with the right amount of theory and practice.
A few weeks ago I finished Martin Ford’s The Rise of the Robots. It is a brilliant book. Sometimes I found the reading a bit heavy because of the sheer amount of details but in general Ford puts here very thought provoking ideas.
He knows the computing world and he knows economic positions left and right and centre and all.
The only part I found week was when he mentioned genetic algorithms. As far as I have seen, that area of computing research has stagnated and will remain so until a new paradigm arises.
The criticisms I read in certain forums seem to be of the usual blokes who deny climate change and who think The Economist is a socialist magazine.
What Ford wrote is linked to Bill Gates’ thoughts on taxing AI and to the very superficial counters about “training will fix it”.
I read an article in Spiegel about a study carried out by Pierre Ibish and his team at the German Institute for Sustainable Development (Hochschule für nachhaltige Entwicklung) Eberswalde. What they did was simply to collect data in order to produce a world view of regions not criss-crossed by roads. That gives us a fascinating but also worrying picture of the Earth. It has been long obvious: roads have a major impact on the environment and the more roads you build through nature, the more stress plants and animals have to go through and the less space we have to breathe clean air.
You can see a more specific map for Germany here.
The thing now would be to sistematically identify areas that could be optimized. We should be thinking now not just about where to prevent road encroachment in nature but actually about what roads can be eliminated for sustainable development.
I am reading Salvanos’ “Professionell entwickeln mit Java EE 7”. The book, like another one by Inden I commented earlier this year is useful for both beginners and experienced Java EE developers, although for these latter much less.
Even if the author wanted to do something for everyone working with Java Enterprise, there was some stuff that he should have taken for granted for someone starting to go into Java EE after having a good grasp of Java’s core. Salvanos spent a lot of space explaining basic stuff on encoding – even making sure you know how to use the right settings in the general Eclipse and Glassfish configurations – and he also bothered to explain very basic topics on relational databases and JDBC .
What I enjoyed was the wealth of details on JSFs and Web services and EJBs. The space and energy Salvanos used for the trivial sections could have been used for more advanced topics on enterprise JavaBeans or RESTFUL services. I suppose you cannot please everyone, even with as Schmöcker, a book of almost 1100 pages.
I haven’t read much fiction lately. Two books stand out: Метро 2033, a Russian science fiction book by Glukhovsky, and Half of a Yellow sun, by Nigerian Chimamanda Ngozi Adichie
The first book was entertaining. The author has a good command of his language. Anyone who has been to the underground in Moscow or Saint Petersburg will see those places differently after reading this novel. Still, I had some issues with the feasibility of the demographics portrayed there. Even if it just a science fiction – or rather a fantastic fiction book, I kept thinking about how realistic it was they would keep such dynamics and that within the area of Moscow’s underground.
I like Adichie’s book much more. It was for me a revelation. There was romance but it was not kitsch. There was a lot of stuff I learnt about Nigeria’s XX century and about the main ethnic groups there. It is, of course, only one instance of an Igbo version. Still, I think Adichie tried to do justice to the different groups.
The book whetted my appetite for more Nigerian literature.
I am reading Hadoop, the Definite Guide, an O’Reilly book written by Tom White. The book is an excellent reference – I have the fourth edition, which is already over one year old -. It gives a short but good introduction of what MapReduce is and then it has a much longer and wonderful course on it all, it goes then into the Hadoop Distributed Filesystem, describes YARN and the I/O. There is a general chapter on Hadoop operations and then it describes briefly but sufficiently soe projects like Sqoop, Flume, Pig, Hive, Crunch, Spark and Zookeeper. Those chapters are rather short, but then you cannot cover it all in one book. The focus is right on what you need to know about Hadoop.
I recently bought Der Weg zum Java-Profi, Konzepte und Techniken für die professionelle Java-Entwicklung, by Michael Inden. As you see from the title, it is a German publication. Its target audience goes from Java developers with at least some experience to experienced developers…yes, all the way to them.
The first 300+ pages offer a brush up of some basic concepts like OOD, inner interfaces and classes, I/O issues and some Java exception handling. Then you have several chapters reviewing topics such as the standard Collections, Guava, log4j and a good overview on multi-threading.
The book then gets more interesting with a bit on reflection, serialization and garbage collection tips. I found the sections on Swing and internationalization a little bit too superficial, though. Perhaps Inden might have just referred to the Oracle stuff we all can see on the Internet for those topics.
From page 795 there is a good introduction on lambda and bulk operations. JavaFX 8 is also presented in a short but decent way. Afterwards you will find chapters on bad smells, refactoring, unit tests and optimization. All in all, there is a bit for everyone. Sometimes the book might come up as superficial but I doubt there is some Java developer who cannot profit from it.
