Skip to content

mayuanyang/doggy

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

29 Commits
 
 
 
 
 
 
 
 

Repository files navigation

doggy

An AI model to classify dog's actions, it uses Keras and Tensorflow to train and Keras-JS with vuejs as the frontend

Setup

Follow this guide to setup the trainning and evaluation environment

Amazon DLAMI

If you are using Amazon DLAMI, all the AI frameworks should be already installed, follow the start up screen to activate the training environment

Your Own Environment

If you are using your own environment (the recommend way is virtualenv), follow this guide to setup Tensorflow https://www.linode.com/docs/applications/remote-desktop/install-vnc-on-ubuntu-16-04

Libraries

We use Keras in this project, follow this guide to install it https://keras.io/#installation

We also use pillow, run the following command to install it

pip install pillow

How to train

Clone the repo

Clone this repo and download the images file

Activate the environment

Activate the TensorFlow, depends on the environment you are in, e.g. for your local own environment, run this command source ~/tensorflow/bin/activate, for DLAMI, run the command for TensorFlow on the welcome screen, usually source activate tensorflow_p36

Start trainning

Depends on where the images are saved, you might need to change the paths a bit inside train.py, change the following as required, by default the dog_dataset folder is the same as the train.py

# data dir
train_data_dir = 'dog_dataset/train'
validation_data_dir = 'dog_dataset/test'

Once the repo and images are downloaded, run this command in the TensorFlow environment python train.py

About

An AI model to classify dog's actions

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published