The official way to save models in a Node environment is tfjs-node.
It also enhances the performance, registrers backend and allows us to use GPUs.
But tfjs-node is not supported on all operating systems, node versions, etc.
Here comes tfjs-node-save - It allows you to use the native filesystem features of tfjs-node, but excludes everything else, so it is supported in any Node environment.
In order tfjs-node-save to work efficiently you should use Node v.8 or higher.**
const tf = rquire("@tensorflow/tfjs");
require("tfjs-node-save");model.save("file://PATH");Refer to the official documentation on how to use the filesystem saving features.
The
file://URL scheme can be used for model saving and loading. For model saving, the scheme is followed by the path to the directory in which the model artifacts are to be saved, for example:
model.save("file://C:/tmp/").then(function() {
console.log("Successfully saved the artifacts.");
});