4

still fairly new to Numpy in Python...I'm trying to build up my own array from RINEX data (see example):

G13 2014 01 02 02 00 00  .440594740212D-04 -.375166564481D-11  .000000000000D+00
      .290000000000D+02  .705937500000D+02  .382980238378D-08 -.135945866650D+01
      .353716313839D-05  .509947887622D-02  .137723982334D-04  .515366394615D+04
      .352800000000D+06 -.819563865662D-07  .312956454846D+01 -.633299350739D-07
      .979542877504D+00  .129500000000D+03  .219020237787D+01 -.757495838456D-08
     -.431803700643D-09  .100000000000D+01  .177300000000D+04  .000000000000D+00
      .240000000000D+01  .000000000000D+00 -.111758708954D-07  .290000000000D+02
      .345618000000D+06  .400000000000D+01

I'm using the following code for initialised matrix:

parameter_block_list = np.empty(cs.TOTAL_SATS, cs.RINEX_NAVIGATION_PARAMETERS) * np.NaN

The problem is the conversion of the numbers in the RINEX file block, where each number uses "D" as an exponential. Is there a way in Numpy to make a data type conversion that caters for such data formats? The error I'm receiving is:

TypeError: data type not understood
Improve this question
3
  • 1
    If you replace the D with an e, would it work?
    – tglaria
    Commented Jan 27, 2016 at 12:27
  • 1
    I think you have two different problems - the error you're seeing is a result of the arguments you've given to np.empty, not the format of your data.
    – Alex Riley
    Commented Jan 27, 2016 at 12:32
  • 1
    I think np.empty takes a tuple of dimensions doesn't it? np.empty((cs.TOTAL_SATS, cs.RINEX_NAVIGATION_PARAMETERS))
    – xnx
    Commented Jan 27, 2016 at 12:33

1 Answer 1

Reset to default
2

I think the reason you're getting data type not understood is that in passing the dimensions of your array to empty as separate arguments, the first is being treated as the shape and the second as the dtype -- see the docs. Perhaps you mean to pass the shape as a tuple:

parameter_block_list = np.empty((cs.TOTAL_SATS, cs.RINEX_NAVIGATION_PARAMETERS)) * np.NaN
Improve this answer
0

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