Skip to content

haraldbo/ikt-450-fish-classification

Repository files navigation

Fish classification using siamese neural networks

This repository contains the code that was developed and used for the IKT450 fish classification project.

Content

siamese.py: Code for training and testing the siamese neural networks.

dataset_functions.py: Functions that were used to download, prepare, and analyze the dataset.

create_report_things.py: Create graphs, confusion matrices, etc. for the report.

dataset.csv: The images that were used, and how they were distributed into train/validation/test sets.

data: Contains statistics that were collected during training and testing.

drawio-diagrams: Contains diagrams that were used in the report.

Examples

Predicted distance between two images

Predicted distance between two images

About

Project: Fish classification using siamese neural networks

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages