Since the introduction of this chapter, you might be wondering that if ML Studio seems so easy and complete, then why does it need extending with coding? If you are thinking so, then let me assure you that this is not the case. To produce a predictive analytics solution for the real world, what ML Studio provides out of the box is quite promising, but very limited. The following are the common scenarios when you may need to write code and integrate with ML Studio:
There is only a limited set of algorithms available through ML Studio. If a certain algorithm is required either for prediction or evaluation, you need to code and integrate the test. For example, there is no specific algorithm available for time series analysis in ML Studio so far.
Though there are some options available, most of the cases of ML Studio with out-of-the box modules are not sufficient to meet the need of exploration and data preparation, which includes data wrangling and...