Skip to content

Multilabel classification on a subset of COCO captions using BiLSTM, CNN and Faster-RCNN components.

Notifications You must be signed in to change notification settings

alexmirrington/coco-multilabel-classification

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

52 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

COCO Multilabel Classification

Running Instructions

Download the dataset, unzip it and place it in the input folder as follows:

|_ code
    |_ algorithm
    |   |_ main.py
    |   |_ ...
    |_ input
    |   |_ data
    |   |   |_ 0.jpg
    |   |   |_ ...
    |   |   |_ 39999.jpg
    |   |_ test.csv
    |   |_ train.csv
    |_ output
    |   |_ predicted_labels.txt
    |_ results
        |_ ...

Install all requirements from requirements.txt:

pip install -r requirements.txt

You can run the deafult configuration for the best model training loop as

python main.py

For further details on the arguments, you can call

python main.py -h

Also, for detail on the arguments you can see the report.

Development Environment Setup

Before contributing to the code base, ensure you have the pre-commit hook set up properly by running the following command:

pre-commit install

This ensures that all committed code adheres to flake8, pylint and pydocstyle linter rules. Additionally, I would recommend installing flake8, pylint and pydocstyle so you can minimise warnings and errors in your text editor or IDE while programming so there is less to fix up before committing.

Running Unit Tests

You can run unit tests with the following command when in the root directory of the project:

python -m unittest discover -s ./code -p "*_test.py"

About

Multilabel classification on a subset of COCO captions using BiLSTM, CNN and Faster-RCNN components.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published