Skip to content

Instantly share code, notes, and snippets.

@metasim

metasim/zonal-stats.ipynb

Created September 10, 2019 22:04
Show Gist options
  • Star 1 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save metasim/76d11b89ae14524ceb75f16ce5e6973d to your computer and use it in GitHub Desktop.
Save metasim/76d11b89ae14524ceb75f16ce5e6973d to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
zonal-stats.ipynb
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Zonal Statistics in RasterFrames\n",
"\n",
"How to roll your own Zonal Stats from existing raster functions\n",
"\n",
"### Import and get a local spark session"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"'0.8.1'"
]
},
"execution_count": 1,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import pyrasterframes\n",
"pyrasterframes.__version__"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"import pyrasterframes.rf_ipython\n",
"from pyrasterframes.rasterfunctions import *\n",
"\n",
"from pyspark import SparkFiles\n",
"\n",
"spark = pyrasterframes.get_spark_session('local[*]')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Vector data\n",
"\n",
"Get a handle to a small [geojson file](https://github.com/locationtech/rasterframes/blob/develop/pyrasterframes/src/test/resources/luray-labels.geojson). "
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"labels_fname = 'luray-labels.geojson'\n",
"labels_url = 'https://raw.githubusercontent.com/locationtech/rasterframes/develop/pyrasterframes/src/test/resources/' + labels_fname\n",
"spark.sparkContext.addFile(labels_url)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Create spark DataFrame of the GeoJSON."
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<table>\n",
"<thead>\n",
"<tr><th>id</th><th>count</th></tr>\n",
"</thead>\n",
"<tbody>\n",
"<tr><td>1</td><td>2</td></tr>\n",
"<tr><td>3</td><td>3</td></tr>\n",
"<tr><td>2</td><td>6</td></tr>\n",
"</tbody>\n",
"</table>"
],
"text/markdown": [
"| id | count |\n",
"|---|---|\n",
"| 1 | 2 |\n",
"| 3 | 3 |\n",
"| 2 | 6 |"
],
"text/plain": [
"DataFrame[id: bigint, count: bigint]"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"labels_df = spark.read.geojson(SparkFiles.get(labels_fname))\n",
"labels_df.groupBy('id').count()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Raster data\n",
"\n",
"Here is a low cloud Landsat8 scene we can use. We will get a RasterFrame with it alongside the vector DataFrame."
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"l8_uri = 'https://landsat-pds.s3.us-west-2.amazonaws.com/c1/L8/015/033/LC08_L1TP_015033_20190727_20190801_01_T1/LC08_L1TP_015033_20190727_20190801_01_T1_B4.TIF'"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"from pyspark.sql.functions import lit\n",
"df = spark.read.raster(catalog=labels_df.withColumn('B4', lit(l8_uri)), \n",
" catalog_col_names=['B4'],\n",
" tile_dimensions=(64,64)) # this is mostly to help visualize things later"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"root\n",
" |-- B4_path: string (nullable = false)\n",
" |-- B4: struct (nullable = true)\n",
" | |-- tile_context: struct (nullable = false)\n",
" | | |-- extent: struct (nullable = false)\n",
" | | | |-- xmin: double (nullable = false)\n",
" | | | |-- ymin: double (nullable = false)\n",
" | | | |-- xmax: double (nullable = false)\n",
" | | | |-- ymax: double (nullable = false)\n",
" | | |-- crs: struct (nullable = false)\n",
" | | | |-- crsProj4: string (nullable = false)\n",
" | |-- tile: tile (nullable = false)\n",
" |-- geometry: geometry (nullable = true)\n",
" |-- id: long (nullable = true)\n",
"\n"
]
}
],
"source": [
"df.printSchema()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Reproject the geometry to the landsat scene's CRS, and filter for only intersecting geometries. This uses GeoMesa's Spark UDT st_intersects."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"df_filtered = df.withColumn('geo_native', \n",
" st_reproject(df.geometry, rf_mk_crs('EPSG:4326'), rf_crs('B4'))\n",
" ) \\\n",
" .filter(st_intersects(rf_geometry('B4'), 'geo_native'))"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"32"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df_filtered.count()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now we have a Spark Dataframe with raster tiles, vector geometry (reprojected), an attribute of the vector data called `id`.\n",
"\n",
"## Zonal Stats\n",
"\n",
"Compute an example zonal statistic: the mean raster value per `id`. \n",
"\n",
"To do the zonal stats, first we will \"burn in\" the vector data. \n",
"\n",
"Then we will mask the raster data."
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
"df_rasterized = df_filtered \\\n",
" .withColumn('raster_id', \n",
" rf_rasterize('geo_native', rf_geometry('B4'), 'id', rf_dimensions('B4').cols, rf_dimensions('B4').rows)) \\\n",
" .withColumn('masked', rf_mask(rf_with_no_data('B4', 0), 'raster_id'))\n",
" "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Take a peek. This will just head the first 5 rows of the spark dataframe. You can double click the tile images to enlarge them in many versions of Jupyter. (And depending on the dimensions of the tile)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<table>\n",
"<caption>Showing only top 5 rows</caption>\n",
"<thead>\n",
"<tr><th>B4</th><th>raster_id</th><th>masked</th></tr>\n",
"</thead>\n",
"<tbody>\n",
"<tr><td><img src=\"data:image/png;base64,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\"></img></td><td><img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAADAFBMVEX95yUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD95yUAAADjX8GUAAABAHRSTlP/AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP8Aao0kNwAAAC9JREFUeAFj+c+AH7Dgl2ZgGFUACaHRcBhJ4cBIaXQzUppxGAlkPaA8ZVaADGAAADHMAYVzQexHAAAAAElFTkSuQmCC\"></img></td><td><img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAYAAABzenr0AAAAmUlEQVR4AWNhGGDAMsD2M4w6YDQERkNgNARGQ2A0BEZDYDQERkNgNARGQ2BAQyDoaOb/AXXAJ5vXA9cvWHrb7H8Ez8WBc8C8ZH8G22WTB8YBHorF/6/FsjI4LyyjvwPsfLr+33RhZhA5ycjw3uk7/R3wi4eJgeMVIwNn+HMGc6Fn9HWAs13r/1cSTAyMZh8ZvKWuMNhw3WIAAPw+IWoblWZqAAAAAElFTkSuQmCC\"></img></td></tr>\n",
"<tr><td><img src=\"data:image/png;base64,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\"></img></td><td><img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAADAFBMVEX95yUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD95yUAAADjX8GUAAABAHRSTlP/AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP8Aao0kNwAAAFBJREFUeAHt0ksKACAIBNARvP+V+2iKELp1o6tgXkYWL9TFdQwM0AnNHBrnQED8yPEtTqRFgQjwxATIm/AfinIRj/D9d2EiBSZy8Pr1Ar3HBjlZCIPQrAUvAAAAAElFTkSuQmCC\"></img></td><td><img src=\"data:image/png;base64,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\"></img></td></tr>\n",
"<tr><td><img src=\"data:image/png;base64,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\"></img></td><td><img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAADAFBMVEX95yUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD95yUAAADjX8GUAAABAHRSTlP/AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP8Aao0kNwAAAF5JREFUeAFj+c+AEzCCZFhwSjOA5fEpgGjFbQLEAApMgBpAvgkwA8g2AW4ADhMQ8jgUIAUv1oBCMoBME8AxDDUGqxVgJ0DSASNuBRCH/iekgExHIgXDqAnQwBgMQQ0AjBgEh8T24Z4AAAAASUVORK5CYII=\"></img></td><td><img src=\"data:image/png;base64,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\"></img></td></tr>\n",
"<tr><td><img src=\"data:image/png;base64,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\"></img></td><td><img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAADAFBMVEX95yUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD95yUAAADjX8GUAAABAHRSTlP/AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP8Aao0kNwAAAGFJREFUeAHtkEsOABAMBbtw/ysjPkV12AsryYzXekHwxELChQsL7SUKdYBgQuco6GowQgMgYXAQdIC/5PTeERaak8ySFhthx4vg4SH4VHdAnBMOrLRhfjE11K5fqEW80UMCJp0Oc8nIqnEAAAAASUVORK5CYII=\"></img></td><td><img src=\"data:image/png;base64,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\"></img></td></tr>\n",
"<tr><td><img src=\"data:image/png;base64,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\"></img></td><td><img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAADAFBMVEX95yUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD95yUAAADjX8GUAAABAHRSTlP/AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP8Aao0kNwAAADhJREFUeAFj+c+AH7Dgl2ZgGFUACaHRcBgNB+S8QiA9MOLPOIxAo/CZAJLHowAsjVsBVBq3ArhHADGWAYUsZSo4AAAAAElFTkSuQmCC\"></img></td><td><img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAYAAABzenr0AAAA0UlEQVR4AWNhGGDAMsD2M4w6YDQERkNgNARGQ2A0BEZDYDQERkNgNARGQ2A0BAYsBN48lf6fdC+A/h0TD7GM/7eL1Rje/etmWK+yi74O0Kjp+39T5D+D+oxXDFNc7RlMeO7TxwExJ5L/P2pUY/h57x/DC6W/DC8dxBjubRFjOHvemLYO0Mvv+y/k94Th5nt2Bv4PvxjYPjAw/D3MxfDG+QfDCptZDLGaSbR1gNikYwzWKb8Y1t4zYOAUYGN46MfE0Oy8iqFzdjjDWytuBuWSdwwAzt48DtOBC+cAAAAASUVORK5CYII=\"></img></td></tr>\n",
"</tbody>\n",
"</table>"
],
"text/markdown": [
"\n",
"_Showing only top 5 rows_.\n",
"\n",
"| B4 | raster_id | masked |\n",
"|---|---|---|\n",
"| <img src=\"data:image/png;base64,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\"></img> | <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAADAFBMVEX95yUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD95yUAAADjX8GUAAABAHRSTlP/AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP8Aao0kNwAAAC9JREFUeAFj+c+AH7Dgl2ZgGFUACaHRcBhJ4cBIaXQzUppxGAlkPaA8ZVaADGAAADHMAYVzQexHAAAAAElFTkSuQmCC\"></img> | <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAYAAABzenr0AAAAmUlEQVR4AWNhGGDAMsD2M4w6YDQERkNgNARGQ2A0BEZDYDQERkNgNARGQ2BAQyDoaOb/AXXAJ5vXA9cvWHrb7H8Ez8WBc8C8ZH8G22WTB8YBHorF/6/FsjI4LyyjvwPsfLr+33RhZhA5ycjw3uk7/R3wi4eJgeMVIwNn+HMGc6Fn9HWAs13r/1cSTAyMZh8ZvKWuMNhw3WIAAPw+IWoblWZqAAAAAElFTkSuQmCC\"></img> |\n",
"| <img src=\"data:image/png;base64,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\"></img> | <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAADAFBMVEX95yUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD95yUAAADjX8GUAAABAHRSTlP/AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP8Aao0kNwAAAFBJREFUeAHt0ksKACAIBNARvP+V+2iKELp1o6tgXkYWL9TFdQwM0AnNHBrnQED8yPEtTqRFgQjwxATIm/AfinIRj/D9d2EiBSZy8Pr1Ar3HBjlZCIPQrAUvAAAAAElFTkSuQmCC\"></img> | <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAYAAABzenr0AAADw0lEQVR4Ae2WfUzUZRzAP7+733FvcBwvJ293gJgFhDiK4UFuWiNJmG7VFv7Ri9lgtRIxzFyttWbZSHDWejEqzbWrZnNzUa0luNQNNw2oNCQwLhDkODle75W7+/365b+S5V9u7Z7nj2f7bs/z/Tyf7/fZHpFbPMRbnJ8YQMxAzEDMQMxAzEDMQMxAzMD/w4B7LFNWITAvRVlqcwk388+8KQOV5btlcdpHJMmIODhKxONhoK0MDe8wJUXoDmWS+1mDnJPh4WRl638CuQ6gKqVOlgNBnIeXEw6JCFe1ZK+4wrArBfUzA8iCgOrPMFJ+HtbvQhyzvstrbju5ukkONW8g3iwwmp5FZcVuuaPrlX+FELM/aZbznz1P6Js01qYN8uOESVknUE1NEq28es2myhSPts3EwOtF3F16iUfTe/h2SkVHVzHbKmTS4uY40rgeySqQMCphbZ/Ae6eF2/bsky+99PwNIcTCF4eUg0sw+GY5NlzMrNPMGTGCqnaBgf13sa/KwdHJUpozPuLTvHvo3VVCi34Zoj+KYaWKvr0reKGllZP6CubyIP2szFsnPqd+53bCiVHWdDTJNyqH2Hj2FK7IrwwE0hkJJDOXOI0kC9g7R3B+tZy2hzcgeAPsYhNRi4m4aIi52+NRuo5Ep4R2Jszmi4+zac/3fPBFDQsJKnY8VIfeHMZ2XKR67W/sHc6Ry3KGFzUhtgxXUZXWxyPmczhkO/0fFhJMFgh2pRFuCFB/tJ22jdUIPj/qGT/u1UswuiNMlIro3ZDdNER3ZwHVhX0ckGqUWJiIUaOcocFvUTEWMvNY/xallK8u+jjEUFSkwjBI7U91BF0GTA/OMO8xsqRbjfaCDkd2OZ5VKYQNqYSSIH5MZtwuknUqzOh9Ir98XUD6hQj3W7Zz8KkD1Mc/fQ3MnyGj88AJRxlxa6YXTf53UDyS72DjyzvIGA9T9GYvBcYrHBoqx7XKQrAwQM+QDbNOwFY7xHlnFhqvlqhOZr5hFvmiApYAo/eqEQwhjMIC/U+8x5kg9ARz+cFdSIOtg3Ktl8R/QBA3r99CyswIhEK0nyvhuKuUYKoENgmdLkJW0gya03r8/ZmYVmoxuiTi5gVmrXqiRomdNe1MhBOxxk1RHAcqZZZpIyzT9JGs9vLGticJP6eooHlRBFGIKMm0yk5R5I6Pffy+Vcc6pZ6nL+cRUG4bTNBwuTYVWQ3GMfBmqAjYfWi7jQilXnq92ZSb/uDLdXYsnR1KYg/N4w/ws6MI/2of7799kK2H61m6v1V2NjZd14h/AXWObirMNJUTAAAAAElFTkSuQmCC\"></img> |\n",
"| <img src=\"data:image/png;base64,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\"></img> | <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAADAFBMVEX95yUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD95yUAAADjX8GUAAABAHRSTlP/AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP8Aao0kNwAAAF5JREFUeAFj+c+AEzCCZFhwSjOA5fEpgGjFbQLEAApMgBpAvgkwA8g2AW4ADhMQ8jgUIAUv1oBCMoBME8AxDDUGqxVgJ0DSASNuBRCH/iekgExHIgXDqAnQwBgMQQ0AjBgEh8T24Z4AAAAASUVORK5CYII=\"></img> | <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAYAAABzenr0AAACX0lEQVR4Ae3V20vTYRzH8fdvzW3+zNU07WDmoVo2sIkJXRjZwUztBBFRkGhLKFkFXZTRhXVTECwJQjpAdGFBBxMVCTHNoSVCF5a1pBTKVbKVh5G6zVz7NS3BP2BbN7/n5nl4br6v5/OclISx7ekwS79MUdj3LeHasduUth1BGa76+fpy6U1fDNnVfXwfUeLyi8R3KMMHwOOl0NiDWuHDbGjlhmMrQ5lSeAA7dKVSQq2Lj6cNbKrq4sqXQvobV6JRER6AoIqgrT8F/aALo2indsCIagxier2hB+QK+yXJ5yMxboRzzxoo7jChdASWniwRYwvDFthr0ilJ62KV2kmScowTWW0MTUXT8CCb35p5oU0gL+uidHfDTQ70Hmbyfi6+SIFYm4cfGZF0lleyvr0stICflzwUdB9l4S0to0VujMsHcVSmEv9qgl3vDzG/MzJ0gOTqy5LTpqZku5UmMQeFQsLWvBrv3kkUwyLP06ooGDseGoCh7oLk69HMvHE1d7bgXQebk/sZiNPhqE2ipdyCKKi4mvE4NICyNe3ce7QT1bgf96LAQXMLNL8zkPREwbLuT9SbUzFpnWzTuIMPyMuskCytWtZ2DzOZoCX29Th+dQSexVGMLwVLVQNRwhRnnRsp0nUGH+BeEU1KvY+D9VYm/GosTbsJ1EN0QNzLIcwVp4gu/orzaSJvW9KCD3hRd0aY3nyrfvabs5KfcFLyj7oQRBHdQzsfctJB78PlXBB8wGzZuX3Tt+szKNz/Zk1/eyFwU8L2Hc8FzY4/F50X/itgGiID5ATkBOQE5ATkBOQE/gAZDckbS5a54gAAAABJRU5ErkJggg==\"></img> |\n",
"| <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAYAAABzenr0AAAM10lEQVR4ASWXeXhV9Z3GP+fcc/fcJTfJvdkTkrCGhFUEBREEARdc2ro9bh15Kn1GHQfFttN2Zhjr2JHijNY6rUtdqMtgfRRFfQJoLCKiCEUCBDAriSG5NzfL3bdzz3xzPf/c5Tnn993e7/u+R1u15nHD2hsGs0ayvpjeH0JRaQLXLheTDSqKDuVH0qiZPNpEEkNV0d1WDJPC2Cwbv936PG+Fl+DSUux/9WIUQFk9RrE9yeDRKkwpSNdl2L58F5fbh9n48JbCufGGLLMf/hZN7ife7McazhCr0qhoRw72kLMZeLvypHwqo3OtRBZkmPWkQnSGG3sog6UniL2smhptksmMnbO/aaZ6Sx+5X5QRCRXz6KNv8O/qRhLPVbH4mhNsdIzzX+FFuO8d4Ne1+4nqdrbfeCvad5dZKTovWbpNWCcN+VSQYrBEDWLVKokKA8cFhYeXtPHMDddS1Z7EHI4TfcHM5H6Vh5b9gNHni+DecfRnarFbJblYnjs+3EzZEXm+XuEBfzvBvEEo42Ktv5O2iRbadkvX5D8tXaLj7lGJ1inE63UsnjT6BTv6efldaVDcCZENMf70nAQ/liT1y0n6TgSYsWkAZ1mExIIairclOfdTK9Mmc2SLCk2l/JBKpkhByUGlyczWCytoa1+IKa1gHYVUa4bMTB2t9Jg6NQWSgTwzmoYIJ5y0LO3lUHwu3rNyow90XcW2NkTPtBJmPaBK65P0315L3a5hHJEk6WovRacsqNkUWadJQAA5q4y2Cv779hfZ1L+OsX+sINBoYI5NJWnCuW6C4Z4StKLBLNEaC4EvYbi/Ft0GnzU7yFdkyA9aKNswSDxjxfRKCU0DKfI2C9p4An2hTu9CJzU7VM6vt1DSYZBxaxJAJ1WqYQ/rKIaJ3y9fRfdPG8g8kGH68ynUVJbhZR7SvSUYVgE2W0Mor1diieiMXKySd+iYQlY+vfF33FlzO/FXqphsUnALKLVwrJCAEkvScH+EVHO1PGPBlAQ1l5eAyJZA1q5gjkPJu53IX2RqMth6LUTrzThGTLj6BGsl0ihVzuzrCcCKLJ7SGLfXneR80kdOTrli11ZMCQV9g6yenGI+aUX32FHjaZlJHsPrxtYzSiDhRU3nCLe6BKw5tGga60iCjN+BYrcRWVaHr2wCymBCKSFSZyVvhro9eSamSzcr2tVCtt+tLOameUc4ba9g28mr0eIKKzZ8Q3ekhK3T2vjVp/fgPplGyQqqbDJgqTRdVyKHKdjPDlCayjA514d3JErXHWV4ugz6N0yjeVEv3W0NMncw5aG0I0eizIQtmMDpcqElZc+DG9P4fWF+0nk799R/LquUpXpVNwf6GrmsvpuUpOzpzZCucgvQ5BQBmeXbYaxjkQKBoQraQ+N4j+aIzyrFHJVgGZi5vZeO39SgzE7R9KwuQLfJ80YBHxmfHdtYFs12bZC7ao9SZwlhU7PE81ay+0o5sbCI21qP8E5PK3/bPw9fwMDzbQzTqJwu+5v3exle7qVyVxdosnrS7mhLGeE5JqlK8ornSc+SNcioFHWYGZ0nU/g6KiAUcDsF1HLO2fsD0um9fl5fs5in5rzJg523EHBGSZSDo8PGRwdXoDkEWE6I1ir4DscIrayQdiqCdIkZhKGbm0jKd1UmkyrXWbnwJFvK97Jx7wOUH7BQ+YnB8HVJPJ/bJWiE4fVVjC3Osra1G38+jJbxyiwTdu4+8mMeW/Auvzh6Q2GlfEeFLXI6Ob8bbWSSXMBDzx0VWBeMc2fTl7zx5PoC42VcKs4hhDEliYzC2aebua16rrAchOdCxSEd3yc2wqtSpIurWH79cdpONPPZR/MoXhJEyxYJ27UVMcUHj9z2w0LlMemc968DBYJSzw+SWzSboeUOVlz1DR9/0cIHz6zG5s4Xdt6cyGP/Lk5wiYd8WEWWiJx0rLpdOONXoyzf0MP+x5cXMDB0GQzcGkB5SKV6+QDRtA1t2vtp4Xwrqm4wa5vwed95DJmxYTJh6EIm8jm8zEmiOk/neABXr8rIRRZsIqCDG4RKnRl87gyRySS5+Pc0rJgNem6wkO8px2LSSdw2QXqPj9j0LMk/5rmp9EvMIrODqWIZgbBXzib8fKSLbMs01P5BCQqx6xfh+uCbAuCqXjxF5MrZ5D/3E1mVY96cfiLCjiVqnu7jNWQPys5Xy3J4RUtGTeiNwh2lknzEgt8W42xfBR6LaMI+E9f8awftoZlUOSaZ5hhFm5LgwM4TKB43Ga+ZzPt16P9XRmlbL0ZjLUpUaHdwCM/xEGfuK8N9RqMz2FhgPXevMFSriJVwfN6po8Ykc7nUPjtZAcF9q/eyq38R1n5LASOTzQZv/8eVuDcP0DFWgdOfRvN1puh8co6IQpZfLtzNa5uFhA4d5fw/L6b2L30y0Bw9/7lEOqGwdtlx/jYxn4vWnubwZ7OZcW8nAzEvsbSVsaBL6FwjMyPJC8teYWfoEp47tRyHPSOrJuQpGpNzig5cYkH5Qw3+B3p558R8NMvAuDidcpa2nOPZJ27ELQwyeqcEf7mL+JJ6kVSVwFcG4zMU9n3RSr4qyz2BAxzJzOHQV7PI2/Nc3nqG00qeUNrH7KoRnh9eySJPP3+3V3N93Ql2lq2k/sM0kQGr4E0Rr5Gl591Gyq4ICWc0lVJ5UOegbzqz3zpJZF0zZX89jSGV2/d34Jhez/Bjhui6CcchD85VYTYduotnbn2RB1+7h1xpms+6G/nT0p08lPsR3aFSIUaDzeWf8kz/Wt786nKcQpijLTZywin2EemGVaHiYJRz031ooVYzta92EdjsYduJdjLGAaabE/x4yQ8EiXHCC7xU3DfI8j3nWCLk0WyZxKqopI08rkWjTEQceFxJQrobnyNBwmwRP5igUovx/rX/Q1usmT++t76AjeIzsl2iIbZRMSPFNqxBYc0D/7QD24Pfgydl6Px6+FK6V4gsJ4cLD5UeCjLt7SD3eI9RanIwKiY1mtfJiiDYzVnWzz7Kpa5z7JucK/J9GK8pzkLrMDZFoUw0QpaZmosGyeZNxAYqmJijM7neoPp5o2AFNYdilmpy/GF8Hq/vvIKAOGAtc7wQfIoDBq8VuRb9/8xTUWBJ50EnH/1sOy7pwquzdvLS+FK2965jpifIfNt56rQ8B1MBLrIG6ckpPH10FUsa+/myqx5jQRbrBTO5rENkWcEh1WhTwZe8vIVpj/2dGuvpKf+FEvCTHw1z3TdDLLI9zc9/spmHb7oZRZtSsjyXvLeFh1d9WCCTy4rOsLuvBbsvi0/ELGso1Gth1h/bROKMF7PQ84mzM6FJHNGrOep3nMYnbqWt5xKCC2UEty68loZkRyFo6oUp/wK3VH1NpXmcbY/dzfsfdGGjny1PnWNPs+j/yvmU1Md48tga9JRWcDXNDd/x7sl53H3Z5wwZZjpSNVg10X1JpuaSAUbfrUEEkt7rbJjiXtq/bsUmgJySAc1IZwqUW/x6hC+/mMW/XP0ODZagzDCLayBLeF0T7b99iv3JYq45Bb9/XxzNt0KzlimbJJ7OkeVUbxVl/klu/GIzjy56r1DE0kA/+w6X0T3ohxl5VLNOw8Jh1FuyGI/Apjs/YlLeDaQEhfTuMnmzCYk9V3ii40o2Np3kQsrN/f/7Jk41LQjPCKonaDCP8rRyFf+24W1ucA5yQc9x9RtbpQp5m6pJM2YUsTu0gNW+Tg68sphctbxTnLFSunqIGZ4Qn7a34rxJwT8nyKl4Jau9neIzXyvjkfo2etN+0necZIZzBKvMMqGbC+YkZVjYlawTwzIqM89x+LYdFCnSAdmC73SXOJzvTef5Y1XoRXmOWypxmVNsue8tXh5Yhk1GUW6PEEoVCQcorPuHQ9TbRrkgPiAq9Kg92/y6AEfl8bMbCqYkatj4Ot7A4qI+zmdKJRELrY4BhOlZYw9RrIrWypUV8FaZouy96wk+STRyONLIx10zsbe7+GTZDAJzI4XVm1s0zETWzmjSybo7DrPI2Vc4q22yGZcvJVQsPxNS5Z55f5b9Fq8nO/7SN8vYNP9zPKZEAekTuoORrIerHENSeV6Cyy7n09RqViElMze7Bgor+One+aKs0FI7xHr3CbrifhYU9fO702uwWXJEsjYORZuIyE02U46uhB9tWNroUDIM6xou8VVNQi47lu5iIFtSSCAhHvFotF5GM8y24KVsLjnAqUyAFusIfgk+dU0l4VPjpANZShaP8bOaD0nJNqwoPsfOwYt5acEr/Lz7xsK9ZpFws+hGkzMoiQgI6wVcU1dQqjyTqWCpkEm9OcyYXlQYQa8YvmrbOKdilZRYEnycmEm5PDP1vvdtNkaj2YEutPxOtIW1807zo5KvZIPEHUviU9ds7whvjl/MxooO+lIl7D7Xgum0vP6tPUexnPf/YPufSfNj7oUAAAAASUVORK5CYII=\"></img> | <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAADAFBMVEX95yUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD95yUAAADjX8GUAAABAHRSTlP/AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP8Aao0kNwAAAGFJREFUeAHtkEsOABAMBbtw/ysjPkV12AsryYzXekHwxELChQsL7SUKdYBgQuco6GowQgMgYXAQdIC/5PTeERaak8ySFhthx4vg4SH4VHdAnBMOrLRhfjE11K5fqEW80UMCJp0Oc8nIqnEAAAAASUVORK5CYII=\"></img> | <img src=\"data:image/png;base64,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\"></img> |\n",
"| <img src=\"data:image/png;base64,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\"></img> | <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAMAAABEpIrGAAADAFBMVEX95yUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAD95yUAAADjX8GUAAABAHRSTlP/AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAP8Aao0kNwAAADhJREFUeAFj+c+AH7Dgl2ZgGFUACaHRcBgNB+S8QiA9MOLPOIxAo/CZAJLHowAsjVsBVBq3ArhHADGWAYUsZSo4AAAAAElFTkSuQmCC\"></img> | <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAYAAABzenr0AAAA0UlEQVR4AWNhGGDAMsD2M4w6YDQERkNgNARGQ2A0BEZDYDQERkNgNARGQ2A0BAYsBN48lf6fdC+A/h0TD7GM/7eL1Rje/etmWK+yi74O0Kjp+39T5D+D+oxXDFNc7RlMeO7TxwExJ5L/P2pUY/h57x/DC6W/DC8dxBjubRFjOHvemLYO0Mvv+y/k94Th5nt2Bv4PvxjYPjAw/D3MxfDG+QfDCptZDLGaSbR1gNikYwzWKb8Y1t4zYOAUYGN46MfE0Oy8iqFzdjjDWytuBuWSdwwAzt48DtOBC+cAAAAASUVORK5CYII=\"></img> |"
],
"text/plain": [
"DataFrame[B4: struct<tile_context:struct<extent:struct<xmin:double,ymin:double,xmax:double,ymax:double>,crs:struct<crsProj4:string>>,tile:udt>, raster_id: udt, masked: struct<tile_context:struct<extent:struct<xmin:double,ymin:double,xmax:double,ymax:double>,crs:struct<crsProj4:string>>,tile:udt>]"
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df_rasterized.select('B4', 'raster_id', 'masked')"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"zonal = df_rasterized.groupBy('id') \\\n",
" .agg(\n",
" rf_agg_mean('masked').alias('mean')\n",
" )"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<table>\n",
"<thead>\n",
"<tr><th>id</th><th>mean</th></tr>\n",
"</thead>\n",
"<tbody>\n",
"<tr><td>3</td><td>9815.90130926921</td></tr>\n",
"<tr><td>2</td><td>7704.144033459678</td></tr>\n",
"</tbody>\n",
"</table>"
],
"text/markdown": [
"| id | mean |\n",
"|---|---|\n",
"| 3 | 9815.90130926921 |\n",
"| 2 | 7704.144033459678 |"
],
"text/plain": [
"DataFrame[id: bigint, mean: double]"
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"zonal"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.7.3"
}
},
"nbformat": 4,
"nbformat_minor": 4
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment