Monitoring with TensorBoard

If you have ever tried to train a NN on your own, then you may know how painful and uncertain it can be. I'm not talking about following the existing tutorials and demos, when all hyperparameters are already tuned for you, but about taking some data and creating something from scratch. Even with modern DL high-level toolkits, where all best practices such as proper weights initialization and optimizers' betas, gammas, and other options are set to sane defaults, and tons of other stuff is hidden under the hood, there are still lots of decisions that you can make, hence lots of things could go wrong. As a result, your network almost never works from the first run and this is something that you should get used to.

Of course, with practice and experience, you'll develop a strong intuition about the possible causes of problems, but intuition needs input data about what's going on inside your network. So you need to be able to peek inside your training...