Closed cmiller8 closed 5 years ago
Hey @talantbekov123 -- I changed the library to scikit-learn instead of shogun
Just pick the most common regression algorithms in scikit-learn and use the same performance metrics
You'll notice that I limited the for
loop through the buildings to only two buildings (for singlebuilding in buildingnames[:2]:
) -- remove the [:2]
and it will cycle through all the buildings
Next step is to combine all of the regression models you tested into that for
loop. Each model should be trained and tested for each building within that loop. Work on ways to create Python functions to make the analysis modular. There will probably need to be a for loop which calls a function to run each model and calculate the metrics.
This means that there will only be one notebook instead of a library of notebooks for each model.
Make sense?
Tested and implemented in the master
notebook
Here is an ML example from start to finish - it has a lot of details about the data set, the test/train data sets: https://www.kaggle.com/claytonmiller/office-building-energy-prediction-random-forest
Replicate this model using alll the regression models in this library: http://scikit-learn.org/stable/supervised_learning.html
You can do this using Python files, Jupyter notebooks, Kaggle Kernels -- how ever you'd like