Community

Using tensorflow machine learning neural network to predict a numeric outcome from numeric inputs

I have a dataset with 3 numeric inputs and 1 numeric output. I want to create a machine learned neural net fit in tensorflow (like I can do in matlab) so that when the user gives 3 numeric inputs to the neural network, it predicts the 1 numeric output. Sounds simple but the data is not linear in nature and in matlab it required training algorithms such as "Levinberg-Marquardt" and others which exist in matlab's 'neural net fitting' app. How can I do this in tensorflow? Is it even possible? I need help getting started. Thanks.