Skip to content
A toy example to identify dog's breed
Python Shell Jupyter Notebook
Branch: master
Clone or download
Pull request Compare This branch is 78 commits ahead of vykozlov:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
data
docker
docs
dogs_breed_det
models
notebooks
references
reports
.dockerignore
.gitignore
Jenkinsfile
LICENSE
README.md
requirements-dev.txt
requirements.txt
setup.cfg
setup.py
test-requirements.txt
test_environment.py
tox.ini

README.md

Dog's breed detector

Build Status

A test example to identify Dog's breed, "Dogs breed detector", as example for DEEPaaS API.

DEEP Open Catalog entry: DEEP Open Catalog

Dogs breed detector is originally forked from udacity/dogs-project, dataset comes from dog dataset.

The project applies Transfer learning for dog's breed identification, implemented with Tensorflow and Keras:

From a pre-trained model (VGG16 | VGG19 | Resnet50 | InceptionV3 | Xception) the last layer is removed, then a new FC classification layer is added, which is trained. All images first pass through the pre-trained network and converted into the tensor with the shape of the 'before-last' layer of the pre-trained network, into so-called 'bottleneck_features'. These bottleneck_features are used then as input for the FC classification network.

Project Organization

├── LICENSE
├── README.md              <- The top-level README for developers using this project.
├── data                   <- Data placeholde
│
├── docs                   <- A default Sphinx project; see sphinx-doc.org for details
│
├── docker                 <- Directory for Dockerfile(s)
│
├── models                 <- Trained and serialized models, model predictions, or model summaries
│
├── notebooks              <- Jupyter notebooks. Naming convention is a number (for ordering),
│                             the creator's initials (if many user development),
│                             and a short `_` delimited description, e.g.
│                             `1.0-jqp-initial_data_exploration.ipynb`.
│
├── references             <- Data dictionaries, manuals, and all other explanatory materials.
│
├── reports                <- Generated analysis as HTML, PDF, LaTeX, etc.
│   └── figures            <- Generated graphics and figures to be used in reporting
│
├── requirements-dev.txt   <- The requirements file for the development environment
│
├── test-requirements.txt  <- The requirements file for the test environment
│
├── requirements.txt       <- The requirements file for reproducing the analysis environment, e.g.
│                             generated with `pip freeze > requirements.txt`
├── setup.cfg              <- makes project pip installable (pip install -e .) so dogs_breed_det can be imported
├── setup.py               <- makes project pip installable (pip install -e .) so dogs_breed_det can be imported
├── dogs_breed_det    <- Source code for use in this project.
│   ├── __init__.py        <- Makes dogs_breed_det a Python module
│   │
│   ├── dataset            <- Scripts to download or generate data
│   │
│   ├── features           <- Scripts to turn raw data into features for modeling
│   │
│   ├── models             <- Scripts to train models and then use trained models to make
│   │                         predictions
│   │
│   └── tests              <- Scripts to perfrom code testing + pylint script
│   │
│   └── visualization      <- Scripts to create exploratory and results oriented visualizations
│
└── tox.ini                <- tox file with settings for running tox; see tox.testrun.org

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

You can’t perform that action at this time.