Skip to content

GeohashPandas module

GeohashPandas

Source code in vgridpandas/geohashpandas.py 
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
@pd.api.extensions.register_dataframe_accessor("geohash")
class GeohashPandas:
    def __init__(self, df: DataFrame):
        self._df = df

    # geohash API
    # These methods simply mirror the Vgrid geohash API and apply geohash functions to all rows

    def latlon2geohash(
        self,
        resolution: int,
        lat_col: str = "lat",
        lon_col: str = "lon",
        set_index: bool = False,
    ) -> AnyDataFrame:
        """Adds geohash ID to (Geo)DataFrame.

        pd.DataFrame: uses `lat_col` and `lon_col` (default `lat` and `lon`)
        gpd.GeoDataFrame: uses `geometry`

        Assumes coordinates in epsg=4326.

        Parameters
        ----------
        resolution : int
            geohash resolution
        lat_col : str
            Name of the latitude column (if used), default 'lat'
        lon_col : str
            Name of the longitude column (if used), default 'lon'
        set_index : bool
            If True, the columns with geohash ID is set as index, default 'True'

        Returns
        -------
        (Geo)DataFrame with geohash IDs added
        """

        if isinstance(self._df, gpd.GeoDataFrame):
            lons = self._df.geometry.x
            lats = self._df.geometry.y
        else:
            lons = self._df[lon_col]
            lats = self._df[lat_col]

        geohash_ids = [
            latlon_to_geohash(lat, lon, resolution) for lat, lon in zip(lats, lons)
        ]

        geohash_col = GEOHASH_COL
        assign_arg = {geohash_col: geohash_ids, f"{geohash_col}_res": resolution}
        df = self._df.assign(**assign_arg)
        if set_index:
            return df.set_index(geohash_col)
        return df

    def geohash2geo(self, geohash_col: str = None) -> GeoDataFrame:
        """Add geometry with GEOHASH geometry to the DataFrame."""
        if geohash_col is not None:
            if geohash_col not in self._df.columns:
                raise ValueError(f"Column '{geohash_col}' not found in DataFrame")
            ids = self._df[geohash_col]
        else:
            if GEOHASH_COL not in self._df.columns:
                raise ValueError(f"Column '{GEOHASH_COL}' not found in DataFrame")
            ids = self._df[GEOHASH_COL]
        return dggs_ids_to_geodataframe(self._df, ids, geohash_to_geo)

    def polyfill(
        self,
        resolution: int,
        predicate: str = None,
        compact: bool = False,
        explode: bool = False,
    ) -> AnyDataFrame:
        """
        Parameters
        ----------
        resolution : int
            Geohash resolution
        predicate : str, optional
            Spatial predicate to apply ('intersect', 'within', 'centroid_within', 'largest_overlap')
        compact : bool, optional
            Whether to compact the Geohash IDs
        explode : bool
            If True, will explode the resulting list vertically.
            All other columns' values are copied.
            Default: False
        """

        result = self._df.geometry.apply(
            lambda geom: polyfill_row(geom, resolution, predicate, compact)
        )

        if not explode:
            assign_args = {GEOHASH_COL: result}
            return self._df.assign(**assign_args)

        result = result.explode().to_frame(GEOHASH_COL)
        return self._df.join(result)

    def geohashbin(
        self,
        resolution: int,
        stats: str = "count",
        numeric_col: str = None,
        category_col: str = None,
        lat_col: str = "lat",
        lon_col: str = "lon",
    ) -> GeoDataFrame:
        """
        Bin points into geohash cells and compute statistics.
        """
        geohash_col = GEOHASH_COL
        df = self.latlon2geohash(resolution, lat_col, lon_col)
        result = aggregate_bin(df, geohash_col, stats, numeric_col, category_col)
        return result.geohash.geohash2geo(geohash_col=geohash_col)

geohash2geo(geohash_col=None)

Add geometry with GEOHASH geometry to the DataFrame.

Source code in vgridpandas/geohashpandas.py 
154
155
156
157
158
159
160
161
162
163
164
def geohash2geo(self, geohash_col: str = None) -> GeoDataFrame:
    """Add geometry with GEOHASH geometry to the DataFrame."""
    if geohash_col is not None:
        if geohash_col not in self._df.columns:
            raise ValueError(f"Column '{geohash_col}' not found in DataFrame")
        ids = self._df[geohash_col]
    else:
        if GEOHASH_COL not in self._df.columns:
            raise ValueError(f"Column '{GEOHASH_COL}' not found in DataFrame")
        ids = self._df[GEOHASH_COL]
    return dggs_ids_to_geodataframe(self._df, ids, geohash_to_geo)

geohashbin(resolution, stats='count', numeric_col=None, category_col=None, lat_col='lat', lon_col='lon')

Bin points into geohash cells and compute statistics.

Source code in vgridpandas/geohashpandas.py 
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
def geohashbin(
    self,
    resolution: int,
    stats: str = "count",
    numeric_col: str = None,
    category_col: str = None,
    lat_col: str = "lat",
    lon_col: str = "lon",
) -> GeoDataFrame:
    """
    Bin points into geohash cells and compute statistics.
    """
    geohash_col = GEOHASH_COL
    df = self.latlon2geohash(resolution, lat_col, lon_col)
    result = aggregate_bin(df, geohash_col, stats, numeric_col, category_col)
    return result.geohash.geohash2geo(geohash_col=geohash_col)

latlon2geohash(resolution, lat_col='lat', lon_col='lon', set_index=False)

Adds geohash ID to (Geo)DataFrame.

pd.DataFrame: uses lat_col and lon_col (default lat and lon) gpd.GeoDataFrame: uses geometry

Assumes coordinates in epsg=4326.

Parameters

resolution : int geohash resolution lat_col : str Name of the latitude column (if used), default 'lat' lon_col : str Name of the longitude column (if used), default 'lon' set_index : bool If True, the columns with geohash ID is set as index, default 'True'

Returns

(Geo)DataFrame with geohash IDs added

Source code in vgridpandas/geohashpandas.py 
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
def latlon2geohash(
    self,
    resolution: int,
    lat_col: str = "lat",
    lon_col: str = "lon",
    set_index: bool = False,
) -> AnyDataFrame:
    """Adds geohash ID to (Geo)DataFrame.

    pd.DataFrame: uses `lat_col` and `lon_col` (default `lat` and `lon`)
    gpd.GeoDataFrame: uses `geometry`

    Assumes coordinates in epsg=4326.

    Parameters
    ----------
    resolution : int
        geohash resolution
    lat_col : str
        Name of the latitude column (if used), default 'lat'
    lon_col : str
        Name of the longitude column (if used), default 'lon'
    set_index : bool
        If True, the columns with geohash ID is set as index, default 'True'

    Returns
    -------
    (Geo)DataFrame with geohash IDs added
    """

    if isinstance(self._df, gpd.GeoDataFrame):
        lons = self._df.geometry.x
        lats = self._df.geometry.y
    else:
        lons = self._df[lon_col]
        lats = self._df[lat_col]

    geohash_ids = [
        latlon_to_geohash(lat, lon, resolution) for lat, lon in zip(lats, lons)
    ]

    geohash_col = GEOHASH_COL
    assign_arg = {geohash_col: geohash_ids, f"{geohash_col}_res": resolution}
    df = self._df.assign(**assign_arg)
    if set_index:
        return df.set_index(geohash_col)
    return df

polyfill(resolution, predicate=None, compact=False, explode=False)

Parameters

resolution : int Geohash resolution predicate : str, optional Spatial predicate to apply ('intersect', 'within', 'centroid_within', 'largest_overlap') compact : bool, optional Whether to compact the Geohash IDs explode : bool If True, will explode the resulting list vertically. All other columns' values are copied. Default: False

Source code in vgridpandas/geohashpandas.py 
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
def polyfill(
    self,
    resolution: int,
    predicate: str = None,
    compact: bool = False,
    explode: bool = False,
) -> AnyDataFrame:
    """
    Parameters
    ----------
    resolution : int
        Geohash resolution
    predicate : str, optional
        Spatial predicate to apply ('intersect', 'within', 'centroid_within', 'largest_overlap')
    compact : bool, optional
        Whether to compact the Geohash IDs
    explode : bool
        If True, will explode the resulting list vertically.
        All other columns' values are copied.
        Default: False
    """

    result = self._df.geometry.apply(
        lambda geom: polyfill_row(geom, resolution, predicate, compact)
    )

    if not explode:
        assign_args = {GEOHASH_COL: result}
        return self._df.assign(**assign_args)

    result = result.explode().to_frame(GEOHASH_COL)
    return self._df.join(result)

poly2geohash(geometry, resolution, predicate=None, compact=False)

Convert polygon geometries (Polygon, MultiPolygon) to Geohash grid cells.

Parameters:

Name Type Description Default
resolution int

Geohash resolution level [1..10]

 required
geometry Polygon or MultiPolygon

Polygon geometry to convert

 required
predicate str

Spatial predicate to apply ('intersect', 'within', 'centroid_within', 'largest_overlap')

 None

Returns:

Name Type Description
list Set[str]

List of geohash ids intersecting the polygon
Example

from shapely.geometry import Polygon poly = Polygon([(-122.5, 37.7), (-122.3, 37.7), (-122.3, 37.9), (-122.5, 37.9)]) cells = poly2geohash(poly, 10, predicate="intersect", compact=True) len(cells) > 0 True

Source code in vgridpandas/geohashpandas.py 
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
def poly2geohash(
    geometry: MultiPolyOrPoly,
    resolution: int,
    predicate: str = None,
    compact: bool = False,
) -> Set[str]:
    """
    Convert polygon geometries (Polygon, MultiPolygon) to Geohash grid cells.

    Args:
        resolution (int): Geohash resolution level [1..10]
        geometry (shapely.geometry.Polygon or shapely.geometry.MultiPolygon): Polygon geometry to convert
        predicate (str, optional): Spatial predicate to apply ('intersect', 'within', 'centroid_within', 'largest_overlap')

    Returns:
        list: List of geohash ids intersecting the polygon

    Example:
        >>> from shapely.geometry import Polygon
        >>> poly = Polygon([(-122.5, 37.7), (-122.3, 37.7), (-122.3, 37.9), (-122.5, 37.9)])
        >>> cells = poly2geohash(poly, 10, predicate="intersect", compact=True)
        >>> len(cells) > 0
        True
    """

    resolution = validate_geohash_resolution(resolution)
    geohash_ids = []
    if isinstance(geometry, (Polygon, LineString)):
        polys = [geometry]
    elif isinstance(geometry, (MultiPolygon, MultiLineString)):
        polys = list(geometry.geoms)
    else:
        return []

    for poly in polys:
        intersected_geohashes = {
            gh for gh in INITIAL_GEOHASHES if geohash_to_geo(gh).intersects(poly)
        }
        geohashes_bbox = set()
        for gh in intersected_geohashes:
            expand_geohash_bbox(gh, resolution, geohashes_bbox, poly)
        for gh in geohashes_bbox:
            cell_polygon = geohash_to_geo(gh)
            if not check_predicate(cell_polygon, poly, predicate):
                continue
            geohash_ids.append(gh)
    if compact:
        return geohash_compact(geohash_ids)
    return geohash_ids

polyfill_row(geometry, resolution, predicate=None, compact=False)

Return cell ids covering a single row geometry.

Source code in vgridpandas/geohashpandas.py 
85
86
87
88
89
90
91
92
93
94
95
def polyfill_row(geometry, resolution, predicate=None, compact=False) -> list:
    """Return cell ids covering a single row geometry."""
    if isinstance(geometry, (Polygon, MultiPolygon)):
        tokens = set(poly2geohash(geometry, resolution, predicate, compact))
    elif isinstance(geometry, (LineString, MultiLineString)):
        tokens = set(
            poly2geohash(geometry, resolution, predicate="intersect", compact=False)
        )
    else:
        raise TypeError(f"Unknown type {type(geometry)}")
    return list(tokens)