Skip to content

Latest commit

 

History

History

segmentation

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 
 
 

Image segmentation

This directory provides examples and best practices for building image segmentation systems. Our goal is to enable the users to bring their own datasets and train a high-accuracy model easily and quickly.

Image segmentation example

Our implementation uses fastai's UNet model, where the CNN backbone (e.g. ResNet) is pre-trained on ImageNet and hence can be fine-tuned with only small amounts of annotated training examples. A good understanding of image classification concepts, while not necessary, is strongly recommended.

Notebooks

The following notebooks are provided:

Notebook name Description
01_training_introduction.ipynb Notebook to train and evaluate an image segmentation model.
11_exploring_hyperparameters.ipynb Finds optimal model parameters using grid search.

Contribution guidelines

See the contribution guidelines in the root folder.