Image-Recognition

Image-Recognition Project

1. Built several CNN models to predict Image-Recognition
2. Data processing to transform training and testing dataset.

Code and Resourced Used

• Python 3.8
• Packages: pandas, numpy, seaborn, matplotlib, keras
• KerasAPI
• multi class loss
• early stop
• adjust parameter
• Youtube

Data Preprocessing

1. Divided train and test data set by 255 to fit the CNN model
2. Zoom in, rotate, flip, and changed the shear range with the train data set, then fit the testing dataset.