1. Leveraged Python advanced statistical methods including MAE, RMSE, PSNR to rigorously evaluateand compare the performance of two machine learning models against real DDM observations2. Crafted 36 heat maps, 16+ scatterplots, and 6 histograms to effectively visualize model behavior anddiscrepancies, utilized bootstrapping with replacement to enhance the reliability of visual results3. Delivered key insights revealing new Model's superior performance over old Model at extreme windspeeds. Identified areas for neural network optimization, aiding improved hurricane forecastingWhat (ideally) will be its impact on the Math clinic sponsor(Spire Global) other stakeholders?Ideally, we hope that our project helps the sponsor answer their question of how well their machine learning operator performs for minimum and maximum wind speeds. It is our hope that we form a conclusion that is well-suppoprted by our error analysis, and that Spire can utilize that to test their assumption and consider making some changes to their neural network architecture to improve the machine learning model's performance. In addition to that, we hope that our project can give the sponsors an explanation on whether the addition of the feature mentioned above is necessary.