Posts
Katie Minjoo Kim
Cancel

Activation Functions in NN(Neural Networks)

What is activation Function? Basically it converts all the sum of inputs into output. It’s like a convert machine which helps to describe data with nonlinearity Why do we use it? It is used ...

Recommender System (3) Types

(Model based) Latent factor Collaborative Filtering So far, we’ve just used similarity based recommendation system but from now on, we use machine-learning based model! The reason why it is also ...

Gradient Tape(2) Example

This is the combination version of TensorFlow 2 quickstart for experts & Migrate early stopping from the official document from Tensorflow! (These two codes work well individually, but if ...

Shap Explainer for LSTM based model

WHAT is SHAP? SHAP(SHapley Additive exPlanations) values are used to explain the output of any machine learning model. They are based on concepts from cooperative game theory, specifically the Shap...

Gradient Tape VS Model.fit

When we build a deep-learning based architecture, model.compile() and model.fit(), which are the Built-in Solution, are mostly used. we can find model.fit() in Beginners Guide of Tensorflow! The ...