cjsackey / dog-vision-project Goto Github PK

End-toEnd multiclass Dog Breed Classification Project.

This project builds and end-to-end multiclass image classifier using TensorFlow 2.0 and TensoFlow Hub.

1. Problem Identifying the breed of a dog given the image.

2. Data From Kaggle's Dog Breed Data Evaluation Competition: https://www.kaggle.com/c/dog-breed-identification/data

3. Evaluation For each image in the test set, we must predict a probability for each of the different breeds. The file should contain a header and have the following format in the link below:

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

4. Features Some information:

We're dealing with images (unstructured data) therefore Deep Learning or Transfer Learning is preferable There are 120 breeds of dogs (120 different classes). There are around 10,000 images in the training set (these images have lables). There are around 10,000 images in the the test set (these images have no labels).