Inspiration
I was working with Signals and Systems during my course for the same here at IIT Bhubaneswar. There came to know about the marvellous idea of analysing component frequencies. Using Fourier Transform, Discrete Time Fourier Transform, Laplace etc..
As I also was interested in machine learning for a while, I decided to try out if concepts of fourier transform work on extending time series, I kept that idea strayed out, but today I finally had the courage to pickup a dataset and start working on it
What it does
It takes in a time-series, which is basically a series of numbers, and gives out as output an extended version of that series, you can select how precise you want the model to align to... and have predictions checked over matlab
How we built it
This is not technically a product, it is just a notebook that I am using to note down my observations. I used Numpy and MatPlotLib mainly.
Challenges we ran into
Using matplotlib was a bit of hassle, as I didn't have practice, for like, a lot of time
Accomplishments that we're proud of
The signal boasts about 0.5% of relative error, which is a great thing.
Log in or sign up for Devpost to join the conversation.