Flower Project
The following project for the course Data Science by Dr. Karsten Tolle. With this project, we want to demonstrate the performance leap between traditional computer vision models and deep neural networks. Please refer to the Report and Presentation for detailed information.
Goal
This project uses two different machine learning models to learn and classify images from different flowers. It trains and compares convolutional neural networks (CNN) and support vector machines (SVM) to learn features for five different flower categories:
- Daisy
- Dandelion
- Rose
- Sunflower
- Tulip
Data
The models trained on the images from the following Kaggle Datasets:
Training the model
The training was processed with Google Colab to take advantage of the fast computation with GPU.
Results
The following picture displays a small ensemble of the test set (not used to train the models) together with the corresponding classifications.
Availability
Source Code
The source code of the machine learning models is released under the MIT License on GitHub: https://github.com/mansurova/flowers
Report and Presentation
The report (submitted: 26.06.2020) and presentation files (presented: 14.07.2020) are also available under CC-BY and MIT License.
