tfrecord

TensorFlow record (.tfrecord) File I/O for Node

Produce data for your TensorFlow pipelines directly in node.

Requirements

This module uses ES2017's async / await, so it requires node.js 7.6 or above.

While this module will presumably be used to interoperate with TensorFlow, it does not require a working TensorFlow installation.

Usage

const tfrecord = require('tfrecord');

async function writeDemo() {
  const writer = await tfrecord.createWriter('data.tfrecord');

  const example = tfrecord.Example.fromObject({
    features: {
      feature: {
        answer: {  // The feature name.
          int64List: {
            value: [42],  // The feature value.
          },
        },
      },
    },
  });

  await writer.writeExample(example);
  await writer.close();
}

async function readDemo() {
  const reader = await tfrecord.createReader('data.tfrecord');
  let example;
  while (example = await reader.readExample()) {
    console.log('%j', example.toJSON());
  }
  // The reader auto-closes after it reaches the end of the file.
}

async function demo() {
  await writeDemo();
  await readDemo();
}

demo();

Development

Run the following command to populate the pre-generated files. These files are distributed in the npm package, but not checked into the git repository.

scripts/generate.sh

The test data can be regenerated by the following command, which requires a working TensorFlow installation on Python 3.

python3 scripts/write_test_data.py

The test data is in the repository so we don't have to spend the time to install TensorFlow on Travis for every run.