POLLIMAC TF

Pollimac TF stands for "Pollen Image Classifier" using TensorFlow. Pollimac TF is an extension of the original SP of Prof. Joman Encinas. It was created using TensorFlow and its pre-classified image classifier, coupled with NodeJS and VueJS for its system.

Setting the app

1. Install TensorFlow
2. After installing TensorFlow, run the virtual environment upon opening your terminals. Upon running the virtual environment via TensorFlow with the command source activate mycondaenv, it should look like:

(mycondaenv) $         

Note: TensorFlow can now be ran with Docker. Just a thought!

Running the app

1. Open 2 terminals, one for the client, and one for the server.

cd pollimac-tf/client # terminal for the client
cd pollimac-tf/server # terminal for the server

2. Run npm install on client and server. Make sure that the version of Node is version 8 or higher. Note: if you're experiencing an error upon npm install on the server, just use the Node version 8.11.1 (switching/installing Node versions is easy with nvm)

3. After installing the packages, run npm start on the server folder. It should run on port 8081. Remember that this should be ran while inside the virtual environment. The virtual environment is only needed for the server folder.

(mycondaenv) $  npm start

4. After running the server, run npm start on the client. It should run on port 8080.

5. Open your browser and go to localhost:8080.

Flow

1. User uploads an image.
2. The web app will classify the uploaded image using the retrained classifier from Tensor Flow.
3. The web app returns the classified image to the user.