How to upload a csv file in tensorflow.js in angular project?

Question

For my anguar project I'm trying to upload a csv file from my assets folder with tf.data.csv, but the file is not being recognized by the code meaning, that the created Object is empty. Is it even possible to upload a csv via tf.data.csv() from assets? And if yes, how? :)

async loadData(){
  const csvUrl = 'assets/tfjs_model/lsm6dsm.csv' as string;
    const csvDataset = tf.data.csv(csvUrl, 
    {
      hasHeader:false, 
      columnNames:["timestamp", "x-axis","y-axis","z-axis"],
    }
    ) as any;
    console.log(csvDataset);
    return csvDataset
  }

Here is what is being printed in the console:

base: e {size: null, input: e}

columnConfigs: undefined

columnNamesValidated: false

configuredColumnsOnly: undefined

delimWhitespace: false

delimiter: ","

fullColumnNames: (4) ["timestamp", "x-axis", "y-axis", "z-axis"]

hasHeader: false

input: e {url: "assets/tfjs_model/lsm6dsm.csv", fileOptions: {…}}

size: null

proto: t

Ivo Oostwegel · Accepted Answer · 2019-09-22 18:31:46Z

1

Yes of course this is possible. I don't know if you CSV has column headers with "" if so try this

async function example(){

// Import from CSV
const dataSet = tf.data.csv('https://raw.githubusercontent.com/JuliaStats/RDatasets.jl/master/doc/datasets.csv');

// Extract x and y values to plot
const pointsDataSet = dataSet.map(record => ({
    x: record["\"Rows\""],
    y: record["\"Columns\""]
}));

const points = await pointsDataSet.toArray();

console.log(points);
}

example();

answered Sep 22, 2019 at 18:31

Ivo Oostwegel

4685 silver badges22 bronze badges

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

Collectives™ on Stack Overflow

How to upload a csv file in tensorflow.js in angular project?

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related