Skip to content

Latest commit

 

History

History
22 lines (15 loc) · 973 Bytes

README.md

File metadata and controls

22 lines (15 loc) · 973 Bytes

MedMINIST_Demo

Binary/multiple classification using MedMINIST dataset

Binary classification task by MedMINIST-pneumoniamnist dataset

  • Motivation: exploring whether basic machine learning models can competent in image binary classification tasks

  • Dataset: MedMINIST-pneumoniamnist (greyscale image)

  • Model: RandomForestClassifier image

Multiple classification task by MedMINIST-pathmnist dataset

  • Motivation: exploring whether basic machine learning models can competent in image multiple classification tasks

  • Dataset: MedMINIST-pathmnist (RGB image)

  • Model: RandomForestClassifier image

Key points

  • We find that basic models can handle these tasks
  • Feature extraction is important for basic models and we have compared several approaches