This was a project during the course "AI Engineering". We Built a Convolutional Neural Network (CNN) model to classify images from a given dataset into predefined categories/classes.
Task Descriptions and Project Instructions
In this project, we classified images from the animals 10 data set.
- Built a sequential CNN model
- Optimized the model
- Prediction accuracy of on holdout data set: 80,99%
- transfer learning from VGG16
- Deployed on Gradio where user could upload image to predict animal class (link expired after 72 hours)
Here is a short description of the folder and files available on the repository.
- holdout_subset.zip. You can use these images to predict with the model
- 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
- Deploy_gradio.ipynb: notebook to deploy the model to a website by using gradio
Use requirements.txt to install the required packages to run the notebooks.