Projects

  • NatureVision
  • BreedSpotter

NatureVision

The NatureVision project utilizes transfer learning techniques to build a deep learning neural network for accurately classifying images of natural scenes. The project evaluates three pre-trained models and achieves an overall accuracy of 95.65%. It focuses on a dataset provided by Intel, which includes images of buildings, forests, glaciers, mountains, seas, and streets. A deployment site is available for users to upload their own images for classification. The readme provides instructions for running the project and highlights the potential applications in fields such as environmental monitoring, tourism, and image organization.

BreedSpotter

The BreedSpotter project utilizes deep learning techniques and transfer learning to accurately classify dog breeds. By leveraging the Xception model and a comprehensive dataset of dog images, the project aims to develop a powerful neural network capable of identifying the top 10 most populated dog breeds with high accuracy. The project provides a deployment site where users can upload dog images and receive real-time predictions of the respective breed.

About Me

Hello, I'm Emuejevoke Eshemitan, a passionate data professional specializing in Machine Learning Engineering and Data Science. My portfolio showcases projects reflecting my dedication to leveraging data for insightful and innovative solutions.
As a Machine Learning Engineer, I excel in designing cutting-edge models using TensorFlow and scikit-learn, building scalable data pipelines, and deploying them in production environments.
As a Data Scientist, I analyze valuable patterns, develop predictive models, and communicate insights through impactful visualizations.
Let's collaborate on a data-driven journey! Explore my portfolio for more.

Phone

+234 902 436 2357

Location

Lagos, Nigeria