 Hello everyone, I am Devesh, a PLB research scholar at the educational department of IAP Bombay. Today, we will talk about a tool called Orange. Orange is an open source, visual programming software package for data visualization, machine learning and data mining. Orange uses common Python libraries such as NumPy, SciPy and Skykit Learn. You can download Orange by visiting this particular layer. This is the welcome screen of Orange. The white blank space here is called the canvas. These icons on the left are called widgets. These widgets are actually the computational units of Orange. They help in reading the data, processing it and visualizing it and building predictive models. The widgets have input-output channels. You just have drag and drop it over the canvas. So this particular widget has only the output channel. A widget in particular may have input-output or both channels. Now these particular widgets have been mainly classified into five categories. The first category is data. The second category is visualize. It also has a couple of widgets, third being model, fourth is evaluate and fifth is unsupervised. These widgets actually communicate with each other with the help of these input-output channels. Now if we double click the file icon, a dialog box will appear. This gives us a choice to use our own dataset or we can also use the dataset that is provided by Orange. This also provides us a description about the various features that are involved. So these are the names of the features. These are the data types. It can be numeric, categorical and the role. So role is basically it is categorized into three categories. Either it is a feature or it is a metadata or it is the value to be predicted. And these are some of the values that these variables take. Now we are choosing this particular file. So now that particular file has been uploaded. Now let's briefly describe the various categories. So the first category of widgets is data. Widgets in this category allows importing the files and pre-processing of the data. The second category is visualize. This has widgets that gives a pictorial representation of the data. The third is model. In model the widgets provide us access to various machine learning models that we can use for regression and classification tasks. We will touch upon some of these models in the course. The next category is evaluate category. So it helps us in the assessment of the models. That is it gives us an idea about how well our model is performing. This is demonstrated through the use of various metrics that we will talk about in the course in the coming weeks. The last category is the widgets in supervised. So these provides us unsupervised models along with some widgets that help in feature transformation.