https://github.com/pottekkat/dog-eyes

This notebook builds a multiclass image classifier for classifying dog breeds using TensorFlow 2.0 and TensorFlow Hub.
https://github.com/pottekkat/dog-eyes

dog-breed-classifier jupyter-notebook kaggle-competition machine-learning machine-learning-models tensorflow tensorflow-examples tensorflow-tutorials

Last synced: 5 days ago
JSON representation

This notebook builds a multiclass image classifier for classifying dog breeds using TensorFlow 2.0 and TensorFlow Hub.

• Host: GitHub
• URL: https://github.com/pottekkat/dog-eyes
• Owner: pottekkat
• Created: 2020-03-07T09:43:22.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2020-03-07T10:25:06.000Z (over 4 years ago)
• Last Synced: 2024-05-02T05:25:45.593Z (7 months ago)
• Topics: dog-breed-classifier, jupyter-notebook, kaggle-competition, machine-learning, machine-learning-models, tensorflow, tensorflow-examples, tensorflow-tutorials
• Language: Jupyter Notebook
• Size: 4.7 MB
• Stars: 5
• Watchers: 4
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Multiclass Dog Breed Classification

This notebook builds a multiclass image classifier using TensorFlow 2.0 and TensorFlow Hub.

## Problem

Identifying the breed of a dog given an image of a dog.

## Data

The data is from Kaggle's dog breed identification competetion.

https://www.kaggle.com/c/dog-breed-identification/data

## Evaluation

The evaluation is a file with prediction probabilities for each dog breed of each test image.

https://www.kaggle.com/c/dog-breed-identification/overview/evaluation

## Features

Information of the data:

* We have images(unstructured data) so it is best to use deep learning/transfer learning.

* There are 120 breeds of dogs i.e 120 different classes.

* There are around 10,000 images in the training and test sets.

    

![Dog Breeds Data](https://github.com/navendu-pottekkat/dog-eyes/blob/master/data.png)

    

## Results

The following image shows some of the predictions made by the model and its prediction probabilities(These predictions were made after training the model with 10% of the training data).

Score in Kaggle after evaluating the test data: 0.89228

![Result- Prediction probabilities](https://github.com/navendu-pottekkat/dog-eyes/blob/master/predictions.png)