This demo helps to visualize the problem of vanishing gradients (particularly w.r.t to use of different activation functions) using TensorFlow.js and Plotly.js.
- Analyse various activation functions by plotting them.
- Train a fully connected neural network using custom parameters.
- Visualize the vanishing gradients in the nn-architecture printed.
- Opacity of links between different nodes represent magnitude of positive and negative gradients separately.
- Visualise accuracy and Loss curves.
These dependencies are required to be installed.
- Node.js version 8.9 or higher
- Yarn
cd into the directory
If you are using yarn:
cd vanishing_gradients
yarn
yarn watch
The package contains two scripts:
yarn watch: starts a local development HTTP server which watches the filesystem for changes so you can edit the code (JS or HTML) and see changes when you refresh the page immediately.
yarn build: generates a dist/ folder which contains the build artifacts and can be used for deployment.