Spark is a lot of fun, but setting it up over Hadoop is non-trivial, based on the installations I've seen. One has to learn all of its conventions as well, although once you grok it, it does indeed play nicely with skLearn. But Dask is a skill-set in and of itself, almost. With the ability to address the Viya engine in Python, it is easy to mix and match the languages. Most of the less-traditional educational initiatives out there for data science and analytics lean heavily on open source tools because you can point anybody at them and get set up for free in no time, no matter where that person might be or what platform they're using. They can (and frequently do, I think) make partnerships with schools to get their software taught to students, but all the cutting edge stuff that's happening in both academia and industry tends to be open source and decidedly not SAS-based. I think the thing that will drain the SAS market share more than anything else is that it's competing against tools that are free for anybody to learn and experiment with and extend. sentiment analysis, maybe basic stemming and topic modeling, if that. Speaking from my experience, the main factor for the company I worked for at the time was less "we can't find someone with the skills (locally or otherwise) to do the work" and more "we've already built everything on top of SAS and integrating something from outside of that ecosystem would be scary." Even then, we didn't go forward with the text analytics add-ons they offered us it was too transparently a bad deal for the limited functionality they offered. Soon you will not even be able to breathe anymore. Microplastics kill plankton and it goes to the oceans in massive quantities by washing synthetic clothing and by emissions from car tires. Just look at the fact that the amount of oxygen on Earth has more than halved in 100 years. It is not for nothing that capitalism is the most destructive system ever used. I think the hysteria is not exaggerated, and that many of the people under the age of 40 will still be able to see clearly in a few years how badly we have been doing for 100 years. Huge amounts of energy will be used for this. My point is, if Python and Javascript were to be replaced by Common Lisp, much less energy would be needed to cool these data centers. Air conditioning and meat industry are very popular in the U.S., and there are many data centers in the U.S.: It's also not difficult where all that technology is going. still consumes around 25% of the world's energy. So you see on a general scale that China has much higher production. Consider, for example, the fact that more than 70% of solar panels are produced in China, and only 3% in the US. I share your opinion that India and China are very polluting with their production methods and use of coal plants, etc. This is not a study conducted by morons, but by the most highly talented PhD scientists. This study takes into account a rapid switch to electric cars, but comes to the conclusion that only far fewer cars are a solution. Just to give you a perspective, there's a study that comes to the conclusion that the world is going to be destroyed around 2050, just by the car industry. Python is one of the most moronic inventions, and the fact that it's so popular says a lot about how thoughtful and vital the work of most computer engineers is today. I can think of very few Python programs that wouldn't be written faster in Common Lisp, and would also draw a lot less power. In fact, Python programmers are just killing the world, instead of doing something intelligent. What Are the Greenest Programming Languages?. Renewable Energy Alone Can’t Address Data Centers’ Adverse Environmental Impact Then SAS Visual Analytics is another story and if you program for Advanced Filter in SAS Viya, then again it's more or less different systax compared with SAS Guide. Company uses SAS likely for security purposes (need an organization who is responsible for the tool if anything bad happened) Unlike if you know MATLAB or Python, you can easily move to R, or even C/C++ (They're interconnected with each other very well, SAS is a standalone hero) If you're in SAS industry for too long, then it's likely hard to move to other jobs with different tool. The syntax is unique and not transferrable. I likely use SAS mainly for practicing SQL. Unfortunately, proc sql has some different characteristics compared with standard SQL (e.g. Lots of problem can be solved with proc sql. The only problem is we have to memorize weird syntax I wonder how you think SAS compared with other languages, any future. I just moved from Python to SAS for 4 months due to new job requirements.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |