Image Classification with CNN

Build a Convolutional Neural Network (CNN) model to classify images from a given dataset into predefined categories/classes.

Task Descriptions and Project Instructions

Project Results

In this project, we classified images from the animals 10 data set.

• Pre-processed data
• Built a sequential CNN model
• Optimized the model
• Prediction accuracy of: 80,99%

Repository Folders and Files

Here is a short description of the folder and files available on the repository.

Documents

• Presentation
• holdout_subset.zip. You can use these images to predict with the model

Notebooks

• split_validation_set: split the data set to one set for training and testing (90%) and a second one to make predictions (10%)
• model_1.ypynb : The starting point model
• model_optimized_ypnb: The optimized model
• transfer_learning_winner.ipynb: using VGG16 to predict the data set

Installation

Use requirements.txt to install the required packages to run the notebooks.