-
Heatmap
-
Cofusion Matrix and Accuracy
Inspiration
Wanted to work with weather data. Got the idea to predict one of the features using the other features. I went to predict RainTomorrow as it will be of great help to predict the rain before going out the next day.
What it does
The code predicts the RainTomorrow feature using the other available feature in the dataset weather.csv.
How I built it
Used the python libraries - pandas, seaborn, sklearn, numpy, matplotlib. The code first reads the dataset and splits it in a ratio of 75:25 for training and testing respectively. Then the sklearn Logistic Regression function is used, a confusion matrix is created and accuracy is calculated using the metrics functions. In the end, a heatmap has been created for a confusion matrix for easy interpretation of accuracy.
Challenges I ran into
I could not use all of the features of the weather dataset because their data types weren't 'integers'. I could not find a way to fit them with the other features. So, I skipped those features and I think that affected the accuracy levels of the model.
Accomplishments that I am proud of
- Successfully completing the training and testing without errors.
- Being able to print the confusion matrix in the heatmap.
What we learned
- Learned how to use Logistic Regression with multiple variables.
What's next for RainTomorrow prediction
I would want to create a live Rain prediction application for a specific place using the weather API.
Log in or sign up for Devpost to join the conversation.