https://github.com/m-romanenko/dog-project

Dog breed classifier
https://github.com/m-romanenko/dog-project

cnn keras python transfer-learning xception

Last synced: about 1 month ago
JSON representation

Dog breed classifier

• Host: GitHub
• URL: https://github.com/m-romanenko/dog-project
• Owner: m-romanenko
• License: mit
• Created: 2018-11-01T10:31:55.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2019-01-12T14:47:07.000Z (about 6 years ago)
• Last Synced: 2024-11-16T09:49:41.263Z (3 months ago)
• Topics: cnn, keras, python, transfer-learning, xception
• Language: Jupyter Notebook
• Homepage:
• Size: 55.6 MB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.txt
  • Codeowners: CODEOWNERS

Awesome Lists containing this project

README

        
[//]: # (Image References)

[image1]: ./my_images/example_brittany.PNG "Example Output: Brittany"

[image2]: ./my_images/example_cat.PNG "Example Output: Cat"

## Dog Classification Project

Given an image of a dog, this algorithm will identify an estimate of the canine’s breed. If supplied an image of a human, the code will identify the resembling dog breed.  

![Sample Output: Brittany][image1]

If supplied an image of something other than a dog or a human, the algorithm will know that, too:

![Sample Output: Cat][image2]

### Instructions

1. The dog dataset can be downloaded [here](https://s3-us-west-1.amazonaws.com/udacity-aind/dog-project/dogImages.zip). You should then extract it to `path/to/dog-project/dogImages`. 

2. The human dataset can be downloaded [here](https://s3-us-west-1.amazonaws.com/udacity-aind/dog-project/lfw.zip). You should then extract it at location `path/to/dog-project/lfw`. 

3. Finally, download the [VGG-16 bottleneck features](https://s3-us-west-1.amazonaws.com/udacity-aind/dog-project/DogVGG16Data.npz) for the dog dataset and place it to `path/to/dog-project/bottleneck_features`.

4. You can now run the Jupyter notebook [dog_app.ipynb](./dog_app.ipynb) and train the model!