FastAI model for detecting water bodies in an image.
Jupyter Notebook version can be used either by training or by exporting pickled data. Install the requirements before executing the file, which is done either by using the command :
pip install fastai wandb opencv-python --user
or by using requirements.txt
file.
Copy or move the required input images to the images
folder next to the notebook file.
Training can be done using an image dataset. Here is an example of an image dataset.