OS X에서 Visual Studio Code를 이용하여 C++코드를 작성하고, 디버깅 하는 방법을 정리해보고자 한다.

공식 사이트 에서 다운로드 후 설치 가능하다.

VisualStudioForMac을 설치해봤지만…현재 Mac Version은 C++ 디버깅을 지원하지 않는다고 한다.

After building your model quickly, how can you improve your model?

What if your model performance is lower than your expectation.

Can you know the problem your model have?

Error analysis is a way to address your model to improve performance.

Before more advanced deep learning algorithm, Let’s look at the way to debugging your NN model.

There are something keeping in your mind to train the model.

Obviously, It is depending on your goal you want to achieve.

Let’s see how you can diagnose your model with some debugging ideas.

In this post, I will introduce the way to speed up training for Neural network with batch normalization.

Normalization is helpful to be converged with gradient descent by reducing the oscillation error.

So, how we can it add into Neural network layer called Batch Normalization Layer

One of the most important part of deep learning is to find the proper weights of model.

In this post we will discuss the ways of optimization for training process.

There are many ways to train with gradient descent.

We’ve studied about the process of traning models. How to train and test our model with data?

How do we evaluate our model after training?

In this post, let’s see the way of training and testing to improve your model performance as you expect.

from the previous post, we can study how to train neural network from loss function with back propagation.

In this post, we will study about the different types of neural network with layer sizes.

The simple meaning of deep neural network is the neural network with many hidden layers.

In the previous post, the topic is about the how neural networks work. In this post, i will introduce the way to learn the neural network.

In learning process, we will compute the cost function.