This is a tensorflow and keras re-implementation of DogBreed_gluon, a code for Dog Breed Identification competition.
Thanks for the author @ypwhs.
- OS: Ubuntu 16.04
- Python: 2.7
- pip: tqdm, numpy, tensorflow(1.4), keras(2.0), pandas, h5py, sklearn
- Place the data files and Stanford Dogs Dataset into root fold.
- Download tensorflow model zoo fold into root fold.
Dog-Breed-Identification
| README.md
| get_features.py
| pred.py
| sample_submission.csv
|
|_______slim
| | setup.py
| | slim_walkthrough.ipynb
| | ...
| |
| |____nets
| | resnet_v1.py
| | inception.py
| | ...
| |
| |____preprocessing
| | vgg_preprocessing.py
| | inception_preprocessing.py
| | ...
| |
| |...
|
|______Images
| |____ n02085620-Chihuahua
| |____ n02085782-Japanese_spaniel
| ...
|
|______test
python get_features.py
python pred.py