Underfitting vs Overfitting
👉🏻 Get Our Complete Data Science Training at 57% OFF: https://bit.ly/3GJOwMR
Download Our Free Data Science Career Guide: https://bit.ly/3QlcVvo
One of the most commonly asked questions at data science interviews is about overfitting. A recruiter will probably bring up the topic and will ask you: “What is overfitting and how do we deal with it?” Fortunately, in this 365 Data Science Tutorial video we will address this issue and prepare you for it. There are two concepts that are interrelated: underfitting and overfitting. They go together and understanding one helps us understand the other and vice versa.
Broadly speaking, overfitting means our training has focused on the particular dataset so much that it has “missed the point” . Underfitting, in the other hand, means that the model has not captured the underlying logic of the data. It does not know what to do and therefore provides an answer that is far from correct. Check out our graph in the video, as it is more intuitive and cooler. First, we will look at a regression problem and identify what a good model, an underfitted and an overfitted model looks like.
Underfitting models are clumsy, have high costs in terms of high loss functions, and their accuracy is low. There is either no relationship to be found or a more complex model is required.
Overfitting, as mentioned earlier, have the opposite problem. They are so good at modelling the data that they miss the point and catch random noise. An example would be, a model that tries to predict the Eurodollar exchange rate, based on 50 common indicators. You train your model that achieves low costs and high accuracies. In fact, you believe with 99,99% you can predict the exchange rate, only to lose all your money in the end. Instead of finding the dependency between the euro and the dollar, it modelled the noise. The noise consists of random decisions from investors participating in the market at that time.
Watch till the end of the video where we will cover overfitting and underfitting in a classification context.
► Consider hitting the SUBSCRIBE button if you LIKE the content: / @365datasciencetutorials
► VISIT our newly gamified platform: https://bit.ly/3XprT5V
🤝 Connect with us on LinkedIn: https://www.linkedin.com/school/365datasci...
Follow us on Instagram: https://www.instagram.com/365datascience/
365 Data Science is an online educational career website that offers the incredible opportunity to find your way into the data science world no matter your previous knowledge and experience. This is why we have dedicated this channel to those who are completely new and are curious to explore the wonderful world of data science. Once you have built a basic theoretical knowledge you can sign up to our comprehensive curriculum where we have prepared numerous courses that suit the needs of aspiring BI Analysts, Data Analysts and Data Scientists.
#overfitting #classification #regression