Midterm coding exam for the EECE693 Neural Networks course.
This dataset contains images of several people involved or concerned with facing the global COVID-19 pandemic along with images of other celebrities.
The people in question are:
- Fadlo Khuri
- Hamad Hassan
- Marcel Ghanem
- Andrew M. Cuomo
- Anthony Fauci
- Tedros Adhanom
- Donald Trump
- Bill Gates
- Keanu Reeves
- Cate Blanchett
- Samuel L. Jackson
The images were collected by running the name of each person as search query in Google Image Search. All images appearing in the search were downloaded at first. 12 images per person were selected and partitioned as such: 10 for training and 2 for testing.
Each image is named by the query used to find it followed by an ordinality number. When using the data for classification, the label for each image would be its name (with the trailing digits and ".jpg" extension removed).
The data is partitioned into two folder:
- "Training": contains 110 labelled images, 10 per class.
- "Testing": contains 22 labelled images, 2 per class.
This folder contains two subfolders:
- Dataset Creation and Model Development
- Model Implementation
The "model_implementation.py" file contains the Python code that needs to be run to build the model and load its learned weights and to be able to obtain the inference for a given image. It is recommended to simply import this file as a module to use it. This module depends on the following Python libraries being installed:
- Numpy
- Tensorflow version 2 or higher
The "example.ipynb"/"example.html" files show an example of how this model can be deployed and used to classify an image downloaded from the web.
The model weights can be found at: https://drive.google.com/open?id=1OEdcpvH2hv729pupwt80qaZrUKLLxx_G