Demonstrate new and original data in deep learning by consuming big data with original algorithms.
Ice Cream Sales Prediction
|
The food industry always needs to know how much product they are projected to sell on any given day and this program helps them do it.
By taking sales from days with different temperatures it is able to accurately predict how ice cream cones they should sell that day. I created an easy to use UI for the user to simply enter what the outside temperature is that day and voila it gives them a projected number of sales along with a graph. Pandas is used as the data processor and then a combination of seaborn and matplotlib are used for visual representation.
Check out the GITHUB here |
Letter Decipher
|
This project uses an Artificial Neural Network (ANN) to decipher what letter it is being presented with. It gets trained on a large dataset of pictures (data) of handwritten letters to "learn" the subtle differences between them.
Once the data is processed by pandas, it is sent over to the keras algorithms to train. Finally the data is correlated between seaborn and matplotlib for visual representation.
|
Neural Networks
|
Here is a bonus teach-back on how Neural Networks.. work, specifically, how weights work in a neural network.
![]()
|
© JOE PARISIA, 2022 | ALL RIGHTS RESERVED