-
-
Our Website
-
What it does?
-
Input-Output
-
Graphical Representation of Quartile Sales
-
Heat Map for our data
-
Machine Learning Code
Inspiration
The current inventory management practices are laced with challenges and inefficiencies, mostly done manually, and thus takes a lot of time. This creates a risk of unavailability of inventory when needed. Indian Army dedicate their lives to keep our country safe . Our team felt that being in the CSE department , we can dedicate our time in creating a solution for technical hurdles faced by our fellow soldiers.
What it does
AI in Inventory Management model can keep all the track of stocks and supply in the particular inventory, and can also give us a prior warning when the stocks will be empty which allows our fellow soldiers to save time and get supplies at appropriate time.
How we built it
We are creating a time series prediction model for our data. Then we will train our AI model using 3 different algorithms, namely ARIMA, tensor flow DNN and XGBoost, and will check SMAPE score of individual algorithm. SMAPE score generally varies depending upon algorithms and dataset given. The best algorithm will provide us with the least SMAPE score and will be used for prediction. After successfully training our model, we will deploy our model using flask and present it on a webpage made from using HTML,CSS and JS.
Challenges we ran into
The main challenge we faced was to choose the algorithm, and on what grounds we should select it.
Accomplishments that we're proud of
We are proud to be able to provide a solution for Indian Army and create and test different time series models with our dataset.
What we learned
We learned how to choose the important data out of a dataset and how to implement data in a model.
What's next for SB-7
We are working for a better optimized Flask , improve front end, and can use different algorithms .
Log in or sign up for Devpost to join the conversation.