Welcome to the Image Classifier project! This project is part of a Udacity nanodegree program and focuses on leveraging the power of Convolutional Neural Networks (CNNs) to classify images from the 102 Oxford Flower dataset. Our goal is to develop a robust image classifier capable of identifying diverse flower species.
The 102 Oxford Flower dataset is a rich collection of floral images, each representing a different species. Our project harnesses the capabilities of CNNs to process and analyze these images, learning to recognize distinctive features and patterns that define each flower species.
Our project encompasses the following key objectives:
-
Data Preprocessing: We carefully preprocess the dataset, including resizing, augmentation, and normalization, to ensure it is compatible with our CNN model.
-
CNN Model Implementation: We build and train a Convolutional Neural Network capable of classifying flower images into their respective species.
-
Model Evaluation: We assess the model's performance using appropriate metrics to ensure it can accurately classify images.
-
Inference: Our trained model can be used for inference, making predictions on new flower images.
-
Documentation: We provide comprehensive documentation to guide users through the project, understand the code, and reproduce results.
You can interact with this project in the following ways:
-
Clone the Repository:
git clone https://github.com/MischaRauch/ImageClassifier.git
-
Data Exploration: Dive into the dataset by running Jupyter notebooks or scripts in the
notebooks
directory to explore and understand the data. -
Model Training: Explore our CNN model implementation in the
models
directory, and experiment with training the model on your own datasets. -
Inference: Use the trained model to make predictions on your own flower images.
-
Contribute: This project welcomes contributions, whether in the form of code enhancements, bug fixes, or insights on improving the model's performance.
This project is licensed under the MIT License.