Kelvin Solorzano

Hardworking and passionate job seeker with strong organizational and problem solving skills eager to secure internships in technology or software engineering position. Ready to help team achieve company goals.

Fashion MNIST Classification using PyTorch

• Built and trained multiple deep learning models using PyTorch to classify images from the Fashion MNIST dataset.
• Implemented fully connected neural network architectures and convolutional neural networks to improve classification accuracy, utilizing techniques such as ReLU activations, max pooling, and dropout.
• Managed the dataset using torchvision utilities and created custom data loaders to handle batch processing for both training and testing phases.
• Compared the performance of different models and architectures (including fully connected networks and CNNs) by tracking key metrics such as training loss, test loss, and accuracy after each training epoch.
• Visualized sample predictions and implemented a reusable evaluation function for assessing model performance on the test dataset.