Input pipeline - reading CSV file for every row

TensorFlow General Discussion
1

I am woriking on the input pipeline with accelerometer data.
I have a CSV file with following columns, where I have all of the rows:
measurement_time|arm|file_name
Using the input pipeline I am trying to read the contents of the file specified in the file_name column (it is a CSV file).
I defined the following map function

def map_ds(row):
    return row[:2],tf.io.read_file(row[2])

But the file is read in a binary format, how do I convert it to a tensor when the content looks like this:
X,Y,Z
4,3,2
…

3

Hi @Tomasz_Gutowski

Welcome to the TensorFlow Forum!

You can refer to this tf.data: Build TensorFlow input pipelines for consuming CSV file data and before using it for model training.

Please share the minimal reproducible code to replicate and to understand if the issue still persists. Thank you.