The Data SDK contains a variety of endpoints for the Hex Tile system.

Hex Tiles is Unfolded's analytic tiling system for geospatial data that combines the ability of tiling systems to handle massive datasets with the power of the H3 hierarchical grid system, enabling a tabular data view with full data integrity at all resolutions.

Create Hex Tiles

Process a geospatial dataset into Hex Tiles using the Unfolded Data SDK. Pass a source UUID ordataset record to convert to Hex Tiles. Requires either a hex (H3) column or latitude and longitude columns.

# Import the HexTileOutputColumn Model from Unfolded Data SDK
from unfolded.data_sdk.models import HexTileOutputColumnConfig
# Import enum members from Unfolded Data SDK
from unfolded.data_sdk.enums import (
    AggregationMethod,
    Dtype
)
# Create Hex Tiles
hextile_dataset = data_sdk.create_hextile(
    source="tree_data_set_hex10",
    source_hex_column="hex10",
    output_columns=(
      # Use HexTileOutputColumnConfig
      # to format source and target columns
      # aggregation method, and data type
      HexTileOutputColumnConfig(
          source_column="TreeID",
          target_column="TreeID_sum",
          agg_method=AggregationMethod.SUM,
          dtype=Dtype.UINT16
      ),
    )
)

curl -X POST https://api.unfolded.ai/internal/v1/datasets/hextile \
-H 'Authorization: Bearer <token>' \
-H 'Content-Type: application/json' \
--data-raw '{
    "source": "<source_dataset_id>",
    "sourceHexColumn": "hex",
    "sourceTimeColumn": "datestr",
    "timeIntervals": ["DAY"],
    "targetResOffset": 4,
    "outputColumns": [
        {
        "sourceColumn": "metric",
        "targetColumn": "metric_sum",
        "aggMethod": "sum",
        "dtype": "uint16"
        }
    ]
}'

You can also process temporal datasets into Hex Tiles. To do this, you must specify the source time column and time intervals.

# Import TimeInterval enum from Unfolded Data SDK
from unfolded.data_sdk.enums import TimeInterval

# Create Hex Tiles
hextile_dataset_time = data_sdk.create_hextile(
    source='sample_dataset_hex6',
    source_hex_column="hex6",
    source_time_column="time",
    time_intervals=(TimeInterval.DAY,)
)

create_hextile reference

Enrich Dataset with Hex Tiles

You can enrich a dataset with Hex Tiles. Enriching refers to the process of joining Hex Tiles with another datasets to enrich them with contextual information. Requires a dataset to enrich with, which can be a UUID, dataset object, or dataframe.

# Enrich dataset with Hex Tiles
data_sdk.enrich(
    dataset="my-dataset-uuid",
    source_id="my-hex-tile-uuid",
    source_column="some_value",
    lat_column="lat",
    lng_column="lng",
    time_column="date",
)

curl -X POST https://api.unfolded.ai/internal/v1/datasets/hextile \
-H 'Authorization: Bearer <token>' \
-H 'Content-Type: application/json' \
-H 'Accept: text/csv' \
--data-raw '{
    "type": "enrich",
    "input": [
        {
            "type": "dataset",
            "uuid": "my-target-uuid"
        }
    ],
    "sourceId": "my-hex-tile-uuid",
    "sourceColumn": "some_value",
    "timeColumn": "date",
    "targetType": "LATLNG",
    "latColumn": "lat",
    "lngColumn": "lng"
}'

enrich_dataset reference.