(应该没人会看到的吧。。。英语稀烂,锻炼下英语吧。不然真成一个老大难问题了)
Waht is Machine Learning:
ML is a subset of “Artificial Intelligence”, computer perform a task use patterns and inference instead of explicit instructions. Machine learning algorithms build a mathematical model based on sample data.
How to work it out.
This project in CSBP would use "supervised learning" which means we use data in the first 5 years of past 7 year to "train" the algorithm, and then compare the result of algorithm prediction for next 2 years to the data of last 2 years in dataset.
Supervised Learning
Supervised learning build a mathematical model of a set of data that contains both the inputs and the desired outputs.
For example, if the task of a algorithm is determine whether an image contains a car, then the dataset(images) would have iamge with and without the car (input), every image should have a label(output) designating whether it contains a car.
But how to do it, the former group left 3 algorithms for us, they are
Recurrent Neural Network(RNN)( Soil RNN.ipynb):
LightGBM(Soil LightGBM.ipynb):
Xgboost(Soil Xgboost.ipynb):
information about this might be updated in next blog when I figure them out.