I'm trying to convert the latitude and longitude values in UJIIndoorLoc dataset to degrees. The paper mentions that the data is in UTM from WGS84.

Here is an example of the values given:

  • LATITUDE: 4864837.920500003
  • LONGITUDE: -7360.497299999

Upon converting these I get values :

(lat: 43.76251700390833, lng: 2.6974029749144584)

But the result I should be getting should be near Universitat Jaume I in Spain (around 39.992845, -0.068439)

Can someone help me figure out the mapping please ?

  • did you figure this out? I have the same issue. thanks
    – maral
    May 2, 2019 at 19:04
  • This does not really answer the question. If you have a different question, you can ask it by clicking Ask Question. You can also add a bounty to draw more attention to this question once you have enough reputation. - From Review
    – BERA
    May 2, 2019 at 19:49
  • Your LATITUDE and LONGITUDE values aren't latitude and longitude values
    – nmtoken
    Oct 25, 2019 at 13:55

2 Answers 2


It took me some time to figure this out, but the UJIIndoorLoc dataset is not in UTM (as mentioned by OP), but in Pseudo Mercator (EPSG:3857).

This Python snippet reads the dataset and unprojects it:

import pandas as pd
from pyproj import Transformer

transformer = Transformer.from_crs(crs_from=3857, crs_to=4326)

df = pd.read_csv('validationData.csv')
unprojected_df = df.apply(lambda xy: transformer.transform(xy['LONGITUDE'], xy['LATITUDE']), axis=1)

The result is:

0      (39.9927560407408, -0.06751662935071567)
1     (39.99241246128007, -0.06633040780517456)
2     (39.99245921530593, -0.06624448268125044)
3     (39.99243372020682, -0.06616833072143029)
4    (39.992981338661174, -0.06864475617804247)

You can see by csv to map with EPSG:3857 on QGIS https://epsg.io/3857

So you have better to reprojection and saved as .csv


Good luck 😉

enter image description here

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.