Binary .dat file Plotting Column Array values

Question

I have imported an array into my IPython notebook using the following method:

SDSS_local_AGN = np.fromfile('/Users/iMacHome/Downloads/SDSS_local_AGN_Spectra.dat', dtype=float)

The array is of the form:

 SPECOBJID_1         RA          DEC           SUBCLASS ...
 299528159882143744  146.29988   -0.12001413   AGN      ...
 299532283050747904  146.32957   -0.30622363   AGN      ...

Essentially each column has a header, and I now need to plot certain values.

As an example, I want to plot RA against DEC...how would I go about doing this?

Perhaps:

axScatter.plot(SDSS_local_AGN[RA], SDSS_local_AGN[DEC])

np.fromfile(file, dtype=float) (using the default sep='') reads a binary file of floating point values. How can your array have "columns" containing integers (SPECOBJID_1) or strings (SUBCLASS)? — Warren Weckesser
– Warren Weckesser, Commented Feb 2, 2015 at 22:15

Adam Hughes · Accepted Answer · 2015-02-02 22:31:20Z

1

Answer is mistaken, see comments

If you want to access them via name, you should use pandas instead of numpy. In numpy, you need to lookup by index:

plt.scatter(SDSS_local_AGN[1], SDSS_local_AGN[2])

But in pandas, it would be as simple as:

df = read_csv('myfile')
df.plot(kind='scatter', x='RA', y='DEC')

http://pandas.pydata.org/pandas-docs/version/0.15.0/visualization.html#scatter-plot

SDSS_local_AGN['RA'] is a valid operation in pandas, but not in numpy.

PS, since you are working in a Notebook, pandas DataFrames will nicely render as HTML tables, making them much more readable.

edited Feb 2, 2015 at 22:31

answered Feb 2, 2015 at 19:01

Adam Hughes

16.5k14 gold badges100 silver badges140 bronze badges

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

4 Comments

Warren Weckesser Over a year ago

Correction: SDSS_local_AGN['RA'] is valid in numpy, if SDSS_local_AGN is a structured array (see docs.scipy.org/doc/numpy/user/basics.rec.html). If the file was a text file with the layout shown in the question, it could be read into a structured array with, for example, SDSS_local_AGN = np.genfromtxt('filename.dat', names=True). However, it is binary file, so neither np.genfromtxt nor pandas.read_csv will work.

Adam Hughes Over a year ago

I didn't realize this, sorry.

Warren Weckesser Over a year ago

The question is confusing (to me, anyway). I don't see how the binary data read with np.fromfile(...) can have the "columns" shown in the question.

Adam Hughes Over a year ago

I'm wondering if OP used "genfromtxt" and posted "fromfile" instead?

Collectives™ on Stack Overflow

Binary .dat file Plotting Column Array values

1 Answer 1

Answer is mistaken, see comments

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Answer is mistaken, see comments

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related