Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>I think you have hit a problem where a different method and some <em>a priori</em> knowledge of the array would really help. The kind of thing where you have a X probability of finding your answer in the first Y percent of the data. The splitting up the problem with the hope of getting lucky then doing this in python with a nested list comprehension or something. </p> <p>Writing a C function to do this brute force isn't too hard using <a href="http://docs.python.org/library/ctypes.html" rel="noreferrer">ctypes</a> either. </p> <p>The C code I hacked together (index.c):</p> <pre><code>long index(long val, long *data, long length){ long ans, i; for(i=0;i&lt;length;i++){ if (data[i] == val) return(i); } return(-999); } </code></pre> <p>and the python:</p> <pre><code># to compile (mac) # gcc -shared index.c -o index.dylib import ctypes lib = ctypes.CDLL('index.dylib') lib.index.restype = ctypes.c_long lib.index.argtypes = (ctypes.c_long, ctypes.POINTER(ctypes.c_long), ctypes.c_long) import numpy as np np.random.seed(8675309) a = np.random.random_integers(0, 100, 10000) print lib.index(57, a.ctypes.data_as(ctypes.POINTER(ctypes.c_long)), len(a)) </code></pre> <p>and I get 92.</p> <p>Wrap up the python into a proper function and there you go.</p> <p>The C version is a lot (~20x) faster for this seed (warning I am not good with timeit)</p> <pre><code>import timeit t = timeit.Timer('np.where(a==57)[0][0]', 'import numpy as np; np.random.seed(1); a = np.random.random_integers(0, 1000000, 10000000)') t.timeit(100)/100 # 0.09761879920959472 t2 = timeit.Timer('lib.index(57, a.ctypes.data_as(ctypes.POINTER(ctypes.c_long)), len(a))', 'import numpy as np; np.random.seed(1); a = np.random.random_integers(0, 1000000, 10000000); import ctypes; lib = ctypes.CDLL("index.dylib"); lib.index.restype = ctypes.c_long; lib.index.argtypes = (ctypes.c_long, ctypes.POINTER(ctypes.c_long), ctypes.c_long) ') t2.timeit(100)/100 # 0.005288000106811523 </code></pre>
    singulars
    1. This table or related slice is empty.
    1. PONumpy: find first index of value fast
      singulars
      1. PTQuestion
    1. PTAnswer
    1. USBrian Larsen
    1. USBrian Larsen
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    1. COIf the array is doubles (remember python floats are C doubles by default) then you have to think a bit harder as == is not really safe or what you want for floating point values. Also don't forget that it is a really good idea when using ctypes to type your numpy arrays.
      singulars
      1. PO
      1. USBrian Larsen
    2. COThanks @Brian Larsen . I might give it a try. I think it's a trivial feature request for the next numpy revision.
      singulars
      1. PO
      1. UScyborg
    3. COI agree with the feature request idea, that is clearly the best way.
      singulars
      1. PO
      1. USBrian Larsen
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload