Skip to content
/ hera Public
forked from keplr-io/hera

Train/evaluate a Keras model, get metrics streamed to a dashboard in your browser.

Notifications You must be signed in to change notification settings

fdoperezi/hera

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

39 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

#Hera

Train/evaluate a Keras model, get metrics streamed to a dashboard in your browser.

demo

Setting up

Step 1. Plant the spy

Install the package


    pip install heraspy

Initialize the spy

    from heraspy.model import HeraModel

    hera_model = HeraModel(
        {
            'id': 'my-model' # any ID you want to use to identify your model
        },
        {
            # location of the local hera server, out of the box it's the following
            'domain': 'localhost',
            'port': 4000
        }
    )

Plant the spy

A spy admits a Keras callback that you can attach to your experiment. For example

    model.fit(
        X_train, Y_train,
        callbacks=[hera_model.callback]
    )

Step 2. Start the server

Git clone this repository, then run


    cd server
    npm install
    node build/server

Step 3. Start the dashboard


    cd client
    npm install
    npm start

Credits

Aside from the obvious ones:

About

Train/evaluate a Keras model, get metrics streamed to a dashboard in your browser.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • JavaScript 91.6%
  • Python 5.9%
  • CSS 1.9%
  • HTML 0.6%