Machine Learning Frameworks

Machine learning approaches like classifiers, regressions, and 0:00 cluster analysis can help you navigate raw data sets that need customized solutions. 0:04 In computer science and 0:09 software engineering, doing custom work like this is often called low level 0:10 because you're working with the most basic pieces of the problem. 0:15 But most approaches to machine learning can be described using statistical 0:20 analysis and math. 0:24 Which means that some of the components for a spam filter that classifies 0:26 something as spam or not spam can be reapplied to another similar problem, like 0:31 classifying different types of zoo animals based on their physical appearance. 0:37 For this reason, there are many high level machine learning frameworks 0:42 that already have all the low level code written so 0:47 you can just plug your code and your data into pre-existing models. 0:51 Later in this course we're going to use a free one called scikit-learn. 0:56 Let's hop into a web browser and take a quick look at the big 1:01 machine learning frameworks that are already out there. 1:05 First up is scikit-learn. 1:09 Scikit-learn is free to use and it's written in Python. 1:12 It allows you to utilize some of the models we've already discussed 1:17 in just a few lines of code. 1:21 Right on the home page, you can see it features tools for 1:24 classification, regression, clustering and more. 1:28 This is a good choice for many different types of projects, 1:34 especially on very customized problem domains, because it's all open-source. 1:37 Next is TensorFlow. 1:43 TensorFlow is another free open-source Python project with 1:45 a different approach to machine learning that uses what are called Tensors, 1:50 which can be thought of as multi-dimensional arrays. 1:55 Then operations can be performed on Tensors, and 1:59 the results can be visualized in graphs. 2:02 This is somewhat of a departure from the machine learning approaches I mentioned 2:05 previously. 2:09 But it's worth exploring the website and 2:10 understanding what TensorFlow has to offer. 2:13 Next, is Amazon Web Services and their machine learning products, 2:16 which is more of a cloud based platform than a framework. 2:21 If we scroll down a bit there's a section that lists some 2:25 API services and that includes image recognition, 2:30 language translation, tools to create chat bots and more. 2:35 This can be a good choice for many different problems, but especially if your 2:41 problem domain can be neatly categorized into one of these high level APIs. 2:47 IBM Watson offers many high level products and services for machine learning. 2:54 They include common problem domains like chat bots and imagine recognition. 2:59 But also includes some others 3:04 that focus on natural language processing and emotional intent. 3:07 Similar to the others, this is a great choice for 3:12 many different types of problems. 3:15 And it can give you a big head start when you need to 3:17 utilize their higher level features that would be difficult to create on your own. 3:21 There are many more frameworks out there so 3:27 check the notes associated with this video for more resources. 3:30 As you've probably gathered, not all machine learning frameworks are the same. 3:34 And in fact, you could argue that not all of these are really frameworks, 3:38 but also software libraries or cloud based platforms. 3:43 The semantics and scopes of these tools varies but as you gain more experience and 3:48 encounter different types of machine learning problems, 3:53 it's important to know that there are many resources available to you. 3:56