> ## Documentation Index
> Fetch the complete documentation index at: https://docs.wherobots.com/llms.txt
> Use this file to discover all available pages before exploring further.

# GeoDataFrame.from_dict

> Construct a GeoDataFrame from a dictionary.

Construct a GeoDataFrame from a dictionary.

*Added in version 2.0.0.*

## Signature

```python theme={"system"}
GeoDataFrame.from_dict(
    data: dict,
    geometry=None,
    crs: Any | None = None,
    **kwargs,
) -> GeoDataFrame
```

## Parameters

<ParamField path="data" type="dict" required>
  Dictionary of the form `{field: array-like}` or
  `{field: dict}`.
</ParamField>

<ParamField path="geometry" type="str or array - like" default="None">
  If a string, the column to use as the active geometry. If an
  array-like, it is added as the `"geometry"` column.
</ParamField>

<ParamField path="crs" type="Any | None" default="None">
  Coordinate reference system to assign to the resulting frame. The
  coordinates are not transformed.
</ParamField>

<ParamField path="kwargs">
  Additional arguments passed to `pandas.DataFrame.from_dict`,
  such as `orient` or `columns`.
</ParamField>

## Returns

<ResponseField name="return" type="GeoDataFrame" />

<Note>
  This constructor first materializes the dictionary in a local
  GeoPandas GeoDataFrame on the driver, then distributes it with Spark.
  It is intended for small in-memory inputs. For large datasets, create
  a Spark DataFrame or use a distributed Sedona data source instead.

  Once distributed, dictionary values follow Spark schema inference.
  Each column must contain values that Spark can represent with one
  compatible type.
</Note>

## Examples

```python theme={"system"}
>>> from shapely.geometry import Point
>>> from sedona.spark.geopandas import GeoDataFrame
>>> data = {
...     "name": ["first", "second"],
...     "geometry": [Point(1, 2), Point(2, 1)],
... }
>>> GeoDataFrame.from_dict(data, crs="EPSG:4326")
     name     geometry
0   first  POINT (1 2)
1  second  POINT (2 1)
```


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.