Half 3: Uncover how a easy Keras sequential mannequin may be efficient
23 hours in the past
One of many widespread issues in time-series evaluation is lacking knowledge.
As now we have seen in Half 1, easy imputation strategies or regression-based fashions like linear regression and determination timber can get us a great distance.
However what if we have to deal with extra delicate patterns and seize fine-grained fluctuations in complicated time-series knowledge?
On this article, we’ll discover how a Neural Community (NN) can be utilized to impute lacking values.
The strengths of NNs are their functionality to seize nonlinear patterns and interactions in knowledge. Though NNs are normally computationally costly, they’ll provide a really efficient approach to impute lacking time-series knowledge in circumstances the place less complicated fashions fail.
We are going to work with the identical dataset as in Half 1 and Half 2, with 10% values lacking, launched randomly for the mock vitality manufacturing dataset.
Don’t miss out Half 1 of this sequence: