Error when using the code with polygon survey

[Eslam-mohakim]

when trying to use the same code df["geometry"] = df["area"].apply(create_geometry) but for area for Polygon type, it fires this error:
"ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (25,) + inhomogeneous part."

this is the function:
def create_geometry(x): if x == x and isinstance(x, dict) and "rings" in x and isinstance(x["rings"], list) and len(x["rings"]) > 0: return Polygon(coordinates) else: return np.nan

[enzo]

The error message setting an array element with a sequence is generally an indication that you're trying to set an array (or in this case, a column in a DataFrame) with a list or a sequence of values when it's expecting only a single value. A possible problem may be that Polygon(coordinates) is returning a sequence (or list) of values instead of a single value and hence the error. Also, you might want to check that each dict your function checks contains rings that themselves do not contain lists that are sequences. try this: def create_geometry(x): if x == x and isinstance(x, dict) and "rings" in x and isinstance(x["rings"], list) and len(x["rings"]) > 0: # assuming that "rings" is a nested list where each sub-list is a coordinate pair coordinates = [tuple(i) for i in x["rings"]] print(coordinates)# look if the coordinates is neasted list return Polygon(coordinates) else: return np.nan don't forget to import numpy impoert numpy as np Il giorno mar 23 apr 2024 alle ore 09:54 Eslam-mohakim < ***@***.***> ha scritto:

…

when trying to use the same code df["geometry"] = df["area"].apply(create_geometry) but for area for Polygon type, it fires this error: "ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (25,) + inhomogeneous part." this is the function: def create_geometry(x): if x == x and isinstance(x, dict) and "rings" in x and isinstance(x["rings"], list) and len(x["rings"]) > 0: return Polygon(coordinates) else: return np.nan — Reply to this email directly, view it on GitHub <#1>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AA2P5KEJPPBFB6TSO6RARKLY6YHUDAVCNFSM6AAAAABGUKXMZSVHI2DSMVQWIX3LMV43ASLTON2WKOZSGI2TQMRQGU2DAMI> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

-- Enzo Cocca PhD in "Science and Technology for Archaeology and Cultural Heritage" ISMEO International Association of Mediterranean and Oriental Studies mail: ***@***.*** cell: [phone]

[enzo]

Furthermore, you should then apply the create_geometry function instead of:

df["geometry"] = df["location"].apply(
lambda x: Polygon(x["rings"][0])
)
in

df["geometry"] = df["area"].apply(create_geometry)
)