I will write with a concept like
1 hour with 3 bottles. If you look at this, it's OK
I try to get about 5 knocks every day. You can probably go while googled, but I think that it will be the shortest distance if you know it systematically.
If you do the above, you will learn the following.
--Experience of writing Python, creating a model, and predicting --Be aware that you can get and try various data by going to Kaggle's site. ――A feeling of what kind of image an algorithm is --The algorithm solves the optimization problem. Sense --Somehow mechanisms and use cases such as regression and gradient boosting --Data preprocessing --One Hot Encoding, normalization, etc.
-Short movie of activation function -What kind of algorithm is available -Environment for machine learning (example: AWS SageMaker) -Introduction to machine learning that can be used in the field by Google data scientists
--XGBoost example --Hyperparameter tuning items, kernel, etc.
――The number of environments where machine learning is possible is increasing.
--Data collection like MLOps-Processing-Efficiency and automation of flow like learning --AI for system development processes like AI Ops --IoT or device system
Recommended Posts