F
Future and Present Weather Predictor using AI
Using local weather station data of Bhubaneswar from SMHI Open Data to predict temperature for the next 24 hours by using different Machine Learning Algorithms.
Created on 13th December 2020
•
F
Future and Present Weather Predictor using AI
Using local weather station data of Bhubaneswar from SMHI Open Data to predict temperature for the next 24 hours by using different Machine Learning Algorithms.
The problem Future and Present Weather Predictor using AI solves
Based on the dataset of weather conditions, predict whether you will be able to play football on the weekend. Usually, to predict weather forecasting, a massive amount of data is being fed into the algorithm that uses deep learning techniques to learn it and then make predictions based on the past data. However, the trained ML model works on a physics free approach for the forecasting process.
Challenges we ran into
- It is actually very difficult to predict the weather of the day due to variability and unpredictibility yet our machine learning model calculates and predicts average temperture & Humidity of the day.
- The efficiency of algorithm was initially 75% to 80% which was due to underfitting of the model but later we came up with Random Forest Regressor which has an efficiency of 97.2%.
- The Data recieved from the api was not in proper format to be feed to the model so we had cleaned the data and trimmed some of the columns to get the data in our array format.
Discussion
Builders also viewed
See more projects on Devfolio