Skip to content

Latest commit

 

History

History
70 lines (36 loc) · 1.58 KB

README.md

File metadata and controls

70 lines (36 loc) · 1.58 KB

Dog Breeds Classifier

Motivation

The purpose of this project is to use a convolutional neural network (CNN) to predict dog breeds. The main object is to evaluate an image then a prediction of which dog breed the dog is, or which dog breed the human most resembles. If the image is detected as neither a dog nor a human, the classifier will not run.

File Descriptions

  • dog_app.ipynb is a Jupyter notebook, contains the whole project code to create a dog breed classifier.
  • The images folder includes all images used for this project.
  • The saved_models folder contains the models saved during this project

Achievements

During the test phase the model achieve the accuracy: 82.1770%

Model

Model

Model Accuracy

Model Accuracy

Model Loss

Model Loss

Results

To check the results I have used:

3 Images from Humans

Human Test 1

Human Test 2

Human Test 3

6 Images from Dogs

dog Test 1

dog Test 2

dog Test 3

dog Test 4

dog Test 5

dog Test 6

Required Libraries

  • Pandas, NumPy, Scikit-learn, tqdm
  • Matplotlib, ImageFile
  • Keras, cv2
  • Glob library

Acknowledgements

The below links, were very useful for completing the projects,