What kind of dog is that? This will let you find out! This project builds an adorable visual interface and a classification algorithm for dog species classification.
-
Download the dog dataset [dog dataset] (link: https://s3-us-west-1.amazonaws.com/udacity-aind/dog-project/dogImages.zip). Unzip it and place it in
. /dog-project/dogImages. -
Download the dog classification VGG-16 network bottleneck feature dataset (https://s3-us-west-1.amazonaws.com/udacity-aind/dog-project/DogVGG16Data.npz) . Place it at
. /dog-project/bottleneck_features. -
Configure the environment in Python > 3.7, pip install -r requirements/requirements.txt
-
run run_rout.py, wait for initialisation to complete, open http://127.0.0.1:5000
-
Run dog_train to train the model.
-
Remind that we store the model by conversions to TFLite