Skip to main content

Python module for interacting with geohashes

Project description

neathgeohash

Coverage Status travis status

Neathgeohash is a Python module that provides functions for finding an approximation of a Line and an ellipsoid (a surface) with a set of geohashes. Behind the neathgeohash library, there is the power of NumPy. The geohasing implementation is approximately thirty times faster than the pygeohash implementation for large data sets.

The code includes Leonard Norrgård’s Geohash code, with necessary extensions to plot vectors using geohashes as pixels.

The example shows ellipsoid with the ellipse center at 51.00 lattitude and 0 longitude and 305m major and 385m minor

Example 1::

import neathgeohash
ellipse = ngh.Ellipse(51, 0, 305, 385, 0)
ellipse_as_ghl = ellipse.estimate_probability_coverage_with_fixed_geohash_depth()
print(ellipse_as_ghl)

{'gcpfpur': 0.0701562499999992, 'gcpfpuq': 0.05140624999999959, 'gcpfpuw': 0.020625000000000015, 
'gcpfpux': 0.02828125000000002, 'u1040h8': 0.02781250000000002, 'u1040h2': 0.06640624999999928, 
'u1040h0': 0.06609374999999929, 'gcpfpun': 0.048437499999999654, 'gcpfpum': 0.03515624999999994, 
'gcpfput': 0.013906250000000009, 'gcpfpup': 0.060624999999999395, 'gcpfpuj': 0.02734375000000002, 
'gcpfpuv': 0.0031250000000000006, 'gcpfpuy': 0.004843750000000002, 'gcpfpuz': 0.005468750000000002, 
'u1040hb': 0.0029687500000000005, 'u1040hc': 0.00234375, 'u1040h9': 0.021562500000000016, 
'u1040h3': 0.05515624999999951, 'u1040h1': 0.048437499999999654, 'u10405c': 0.018750000000000013, 
'gcpfpgz': 0.022812500000000017, 'gcpfpgv': 0.011093750000000006, 'gcpfpuk': 0.01687500000000001, 
'gcpfpus': 0.008437500000000004, 'gcpfpuh': 0.01484375000000001, 'gcpfpuu': 0.0009375000000000001, 
'u10405b': 0.024062500000000018, 'gcpfpgy': 0.01703125000000001, 'gcpfpgu': 0.0057812500000000025, 
'gcpfpvn': 0.00015625, 'gcpfpvp': 0.00015625, 'u1040j0': 0.000625, 'u1040j1': 0, 'u1040hf': 0.0009375000000000001, 
'u1040hd': 0.01625000000000001, 'u1040h6': 0.03281249999999999, 'u1040h4': 0.028593750000000022, 
'u10405f': 0.010312500000000006, 'u10405d': 0.00109375, 'u104058': 0.0028125000000000003, 
'gcpfpgw': 0.0017187499999999998, 'gcpfpgs': 0.000625, 'gcpfpu7': 0.005937500000000003, 
'gcpfpue': 0.0018749999999999997, 'gcpfpu5': 0.005937500000000003, 'gcpfpug': 0.00046875, 
'gcpfpgg': 0.0020312499999999996, 'u104059': 0.005312500000000002, 'gcpfpgx': 0.003906250000000001, 
'gcpfpgt': 0.0021874999999999998, 'gcpfpge': 0.00015625, 'u1040hg': 0.000625, 'u1040he': 0.006718750000000003, 
'u1040h7': 0.01609375000000001, 'u1040h5': 0.01593750000000001, 'u10405g': 0.004218750000000001, 
'u10405e': 0.00078125, 'u1040hu': 0.00046875, 'u1040hs': 0.00140625, 'u1040hk': 0.005000000000000002, 
'u1040hh': 0.007031250000000004, 'u10405u': 0.0015624999999999999, 'u10405s': 0.00046875}

Example 2::

Example

To see image follow the url: https://github.com/mpdwulit/neathgeohash/blob/master/samples/ellipse_coverage.geojson

Performance ~(30x faster)

In version 0.5.0, the geohash encoding was added. The original code was copied from https://github.com/vinsci/geohash and extended by fast_encode function. Fast_encode for performance uses NumPy. Here is a comparison between initial implementation and NumPy implementation. Implementation based on blog: https://mmcloughlin.com/posts/geohash-assembly

Performance testing code

Performance

=========ngh.encode()================

         63000002 function calls in 26.743 seconds

   Ordered by: cumulative time

   ncalls  tottime  percall  cumtime  percall filename:lineno(function)
        1    0.308    0.308   26.743   26.743 performance_cmp.py:22(<listcomp>)
  1000000   22.988    0.000   26.436    0.000 geohash.py:86(encode)
 61000000    3.255    0.000    3.255    0.000 {built-in method builtins.len}
  1000000    0.193    0.000    0.193    0.000 {method 'join' of 'str' objects}
        1    0.000    0.000    0.000    0.000 {method 'disable' of '_lsprof.Profiler' objects}


=========ngh.fast_encode()===========

        160 function calls (159 primitive calls) in 0.899 seconds

   Ordered by: cumulative time

   ncalls  tottime  percall  cumtime  percall filename:lineno(function)
        1    0.114    0.114    0.899    0.899 geohash.py:167(fast_encode)
    36/35    0.366    0.010    0.366    0.010 {built-in method numpy.core._multiarray_umath.implement_array_function}
       32    0.000    0.000    0.336    0.010 <__array_function__ internals>:2(where)
       26    0.273    0.010    0.273    0.010 {built-in method numpy.array}
        1    0.086    0.086    0.111    0.111 geohash.py:123(__encode_into_uint64)
        1    0.051    0.051    0.066    0.066 geohash.py:149(__encode_base32)
        1    0.000    0.000    0.017    0.017 <__array_function__ internals>:2(dot)
        1    0.000    0.000    0.014    0.014 <__array_function__ internals>:2(column_stack)
        1    0.000    0.000    0.014    0.014 shape_base.py:612(column_stack)
        1    0.000    0.000    0.014    0.014 <__array_function__ internals>:2(concatenate)
        1    0.008    0.008    0.008    0.008 {method 'dot' of 'numpy.ndarray' objects}
       32    0.000    0.000    0.000    0.000 multiarray.py:312(where)
        1    0.000    0.000    0.000    0.000 <__array_function__ internals>:2(squeeze)
        3    0.000    0.000    0.000    0.000 {method 'reshape' of 'numpy.ndarray' objects}
        1    0.000    0.000    0.000    0.000 shape_base.py:608(_column_stack_dispatcher)
        1    0.000    0.000    0.000    0.000 fromnumeric.py:1426(squeeze)
        1    0.000    0.000    0.000    0.000 shape_base.py:209(_arrays_for_stack_dispatcher)
       12    0.000    0.000    0.000    0.000 {method 'append' of 'list' objects}
        1    0.000    0.000    0.000    0.000 {built-in method builtins.hasattr}
        1    0.000    0.000    0.000    0.000 {method 'squeeze' of 'numpy.ndarray' objects}
        1    0.000    0.000    0.000    0.000 {built-in method builtins.isinstance}
        1    0.000    0.000    0.000    0.000 {method 'disable' of '_lsprof.Profiler' objects}
        1    0.000    0.000    0.000    0.000 multiarray.py:707(dot)
        1    0.000    0.000    0.000    0.000 fromnumeric.py:1422(_squeeze_dispatcher)
        1    0.000    0.000    0.000    0.000 multiarray.py:145(concatenate)

TODO

  1. Add more efficient implementation for finding geohash coverage
  2. Add Box-Muller Sampling from multivariate normal distribution
  3. Improve Monte Carlo integration

License

Copyright (c) 2020 Marek Dwulit

MIT License

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Keywords

Geohash, GIS, latitude, longitude, encode, decode, Galileo, GPS, WGS84, coordinates, geotagging.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

neathgeohash-0.6.4.tar.gz (14.1 kB view hashes)

Uploaded Source

Built Distribution

neathgeohash-0.6.4-py2.py3-none-any.whl (14.7 kB view hashes)

Uploaded Python 2 Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page