This repository contains an implementation of an scene classificator in Python using Machine Learning techniques.
The project is in the framework of Master of Computer Vision Barcelona's Module 3.
Place your train and test dataset using this file structure:
- data/train/*/*.jpg
- data/test/*/*.jpg
If using Pipenv add a file called .env with the environment var PYTHONPATH set to the root of this project.
PYTHONPATH=[absolute_path_to_project_root]
if not is possible you have to set it in your environment using,
export PYTHONPATH=[absolute_path_to_project_root]`
The generated report and resources can be found in the folder results of this repository.
The dataset is splitted in several ways, some are redundant:
- train + validation_and_test (1881 + 807)
- train + validation + test (1881+320+487)
- train_small (400)
- train_toy (80)