Use Tensorboard With Tensorflow.js

Question

I would like to use tensorboard to visualize my model fitting.

The instructions for the tensorboard tutorial (which is for python) and in the API docs (JS) use tf.node.

But tf.node does not exist.

TypeError: Cannot read property 'summaryFileWriter' of undefined

My code:

require('@tensorflow/tfjs-node');
console.clear();
require('dotenv').config();
const debugTf = require('debug')('tf');
const debugTfVerbose = require('debug')('tf:verbose');
const tf = require('@tensorflow/tfjs');
const summaryWriter = tf.node.summaryFileWriter('/tmp/tfjs_tb_logdir');

I'm using "@tensorflow/tfjs-node": "^1.2.3"

edkeveked · Accepted Answer · 2019-06-28 09:14:52Z

2

According to the documentation, the namespace tf should be associated to the package tfjs-node instead of tf

const tf = require('@tensorflow/tfjs-node');

answered Jun 28, 2019 at 9:14

edkeveked

18.4k10 gold badges60 silver badges95 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Andi Giga Over a year ago

I always thought the one require is to load the engine e.g. gpu or cpu and the other one is for the library. Was already wondering why I land on to diff. documentations from google. Thx

Collectives™ on Stack Overflow

Use Tensorboard With Tensorflow.js

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related