Benchmarking classical machine learning methods for image classification with popiular deep learning methods. The goal of this project is to explore the capabiliities of non-deep learning methods. Single regression, logistic regression, multiple regression KNN, SVM, decision trees, k-means clustering, random forests, and Naive Bayes algorithm. These methods will be compared with each other and three deep learning methods - . These models represent popular CNN methods, image transformers, and an architecture focused on decreasing CPU power.
Using the BIRDS 525 SPECIES- IMAGE CLASSIFICATION dataset from gpiosenka on Kaggle
Use accuracy - just set up inference folder so that it can take this
[] look into whether using one hot encoding works better or no [] collect time to train and cpu/gpu necessary [] benchmark dataset against popular image class NN and one ViT
look into using tabular data