How to convert date from unix timestamps to human readable format in graphlab (python 2.7)

Question

I am working on large text file of 34gb. I have successfully parsed the file using the graphlab create. There is a column in the dataset about the date. The date is displayed in a unix timestamps. How can I convert a UNIX timestamp from the input file (converted to an SFrame) into a human readable format?

Answer 1

It's a bit quirky, but you can just cast the column of unix timestamps to type "datetime.datetime" (provided you imported datetime).

This code:

import graphlab as gl
import datetime as dt
sf = gl.SFrame({'a':[1,2,3]})
sf['a'] = sf['a'].astype(dt.datetime)

Produces this:

Columns:
    a       datetime

Rows: 3

Data:
+---------------------+
|          a          |
+---------------------+
| 1970-01-01 00:00:01 |
| 1970-01-01 00:00:02 |
| 1970-01-01 00:00:03 |
+---------------------+
[3 rows x 1 columns]