I have a dataset containing 34 input columns and 8 output columns.One way to solve the problem is to take the 34 inputs and build individual regression model for each output column.
I am wondering if this problem can be solved using just one model particularly using Neural Network.
It sounds like a normal multi-dimensional regression problem. Many methods including feed forward neural network should work. Integrated neural networks and deep learning will be able to model more complex problems and provide predictions that resemble the learning processes. Some reference materials https://www.coursera.org/lecture/deep-neural-networks-with-pytorch/multiple-output-linear-regression-training-ByhfU How To Select Neural Networks For Multiple Output Regression In Excel : https://search.yahoo.com/yhs/search?type=mdru03544ed8790c731199f7¶m1=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%3D&hsimp=yhs-001&hspart=rotz&fr=yhs-rotz-001&ei=UTF-8&fr2=rs-algo%2Cp%3As%2Cv%3Aw%2Cm%3Ars_algo&p=how%20to%20select%20neural%20networks%20for%20multiple%20output%20regression%20in%20excel
Can you specify what exactly you mean with output columns?
In the context of NN, I would simply say that your output vector is not a single scalar value, but an 8 dimensional vector. So yes, any NN based model would work.
There are many functions in the python Sklearn library that perform multi-output regression tasks in a few lines of code. I put a link below to a simple step-by-step implementation guide.
https://machinelearningmastery.com/multi-output-regression-models-with-python/
I am not quite clear what you exactly wants to describe. If I understood you correctly, the question is when a linear relationship governs on your data, why you want to use non-linear models such as Neural networks?
Please look at the class
Machine learning
by Professor Yaser
on YouTube
from
Caltech university
You may follow http://oa.upm.es/40804/1/INVE_MEM_2015_204213.pdf
http://matlabtools.com/wp-content/uploads/p603.pdf
https://arxiv.org/pdf/1901.00248.pdf
- Badges
- Science topic