Felix Abbot

Felix’s innovative solution revolves around revolutionising remote sheep monitoring. With a vision for more effective and practical methods, he developed a cutting-edge semi-supervised machine learning project plan. This model assimilates diverse input data, including cloud maps, soil quality, and infrared sheep emissions, to predict future outcomes with remarkable accuracy and precision. 

Felix’s ingenuity lies in his clever manipulation of modern technologies like LiDAR, thermography, and soil maps. These elements were artfully woven together into a Machine Learning project plan featuring ReLU and Softmax Activation functions, Bayesian Optimization, Random Forest, and Cross-entropy loss. He meticulously fine-tuned neural network considerations, including hyperparameters, to optimise performance.  

The model's training grounds were the rich data reserves of the New Zealand Soil Information System, further refined through the use of handheld field spectrometers. Felix’s solution represents a significant leap forward in the realm of remote sheep monitoring, promising more efficient and data-driven management practices for the future.