Skip to content

Simple classification script that uses VGG16. Learned on image-net

Notifications You must be signed in to change notification settings

saworz/object-classification-vgg16

Repository files navigation

Object classification - VGG16

Pretrained model that can be used to classify objects from squirrels up to military planes. Full list of classes: https://image-net.org/challenges/LSVRC/2014/browse-synsets

Project Organization

├── README.md          <- The top-level README for developers using this project.
│
├── requirements.txt   <- The requirements file for reproducing the analysis environment, e.g.
│                         generated with `pip freeze > requirements.txt`
│
├── setup.py           <- makes project pip installable (pip install -e .) so src can be imported
│
├── Animals_classification_VGG16                <- Source code for use in this project.
│   ├── __init__.py    <- Makes src a Python module
│   │
│   ├── data           <- Scripts to handle data
│   │   └── extract_dataset.py <- Unzips dataset
│   │   └── get_images.py      <- Gets list of all .jpeg, .jpg or .png files in dir
│   │   └── parse_data.py      <- Data parser
│   │   └── split_folders.py   <- Splits data into training and validation dirs
│   │
│   │
│   ├── features       <- Scripts to turn raw data into features for modeling
│   │   └── handle_input.py    <- Used to choose between training and loading model
│   │   └── image_folder.py    <- Custom ImageFolder class for data validation
│   │   └── predict_custom_image.py  <- Used to validate model on a custom data
│   │
│   │
│   ├── model        <- Scripts to train/load/save model
│   │   │── create_model.py    <- Initialize new model                
│   │   ├── load_model.py      <- Load existing weights to the model
│   │   └── save_model.py      <- Save trained model's weights to file
│   │   └── training.py        <- Training pipeline
│   │
└── tox.ini            <- tox file with settings for running tox

Project based on the cookiecutter data science project template. #cookiecutterdatascience

About

Simple classification script that uses VGG16. Learned on image-net

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages