dotnet/machinelearning · error · NotSupportedException

String.Format(Microsoft.Data.Strings.VectorSubTypeNotSupport

Error message

String.Format(Microsoft.Data.Strings.VectorSubTypeNotSupported, itemType.ToString())

What it means

GetVectorDataFrame allocates a VBufferDataFrameColumn<T> only for a fixed set of vector item types; for any other itemType it throws NotSupportedException(Strings.VectorSubTypeNotSupported, itemType). Reached from ToDataFrame when a vector column's element type is unsupported.

Source

Thrown at src/Microsoft.Data.Analysis/IDataView.Extension.cs:212

            }
            else if (itemType.RawType == typeof(ushort))
            {
                return new VBufferDataFrameColumn<ushort>(name);
            }
            else if (itemType.RawType == typeof(char))
            {
                return new VBufferDataFrameColumn<char>(name);
            }
            else if (itemType.RawType == typeof(decimal))
            {
                return new VBufferDataFrameColumn<decimal>(name);
            }
            else if (itemType.RawType == typeof(ReadOnlyMemory<char>))
            {
                return new VBufferDataFrameColumn<ReadOnlyMemory<char>>(name);
            }

            throw new NotSupportedException(String.Format(Microsoft.Data.Strings.VectorSubTypeNotSupported, itemType.ToString()));
        }
    }

}

View on GitHub (pinned to 7b76e69cf9)

Solutions

1. Change the pipeline so the vector item type is a supported primitive (e.g. float, double, ReadOnlyMemory<char>).
2. Project the unsupported vector into a supported representation in the source IDataView.
3. Remove or split the unsupported vector column before calling ToDataFrame.
4. Copy the column manually into a VBufferDataFrameColumn<T> of a supported T.

Example fix

// before
var df = dataView.ToDataFrame(); // vector of unsupported item type
// after
// expose the column as a vector of float in the source:
Output = new VectorDataViewType(NumberDataViewType.Single, size);

bool ok = dataView.Schema.Where(s => s.Type is VectorDataViewType v)
    .All(s => IsSupportedItemType(((VectorDataViewType)s.Type).ItemType));

static bool IsSupportedItemType(DataViewType t) =>
    t == NumberDataViewType.Single || t == NumberDataViewType.Double ||
    t == TextDataViewType.Instance || t == NumberDataViewType.Int32;

try { var df = dataView.ToDataFrame(); }
catch (NotSupportedException ex) when (ex.Message.Contains("vector"))
{ /* unsupported vector item type: reproject the column */ }

When it happens

Trigger: Converting an IDataView whose schema has a VectorDataViewType with an item type outside the supported list (e.g. vectors of TimeSpan or unknown structs).

Common situations: Custom IDataView implementations exposing vector columns of unusual element types; transforms producing vectors of non-standard primitives; exotic data sources fed to ToDataFrame.

Understand the failure class

Background: UnsupportedOperationException and "is not supported" errors: when a library deliberately refuses a call — this error's family across 30 libraries.

Related errors

• String.Format(Microsoft.Data.Strings.NotSupportedColumnType,
• Validation data has 0 rows
• Training data and validation data schemas do not match. Trai
• Training data and validation data schemas do not match. Colu
• Training data and validation data schemas do not match. Colu

AI-assisted analysis of dotnet/machinelearning@7b76e69cf9 (2026-09-11). Data as JSON: /api/errors/653c933e11ec2cf9. Report an issue: GitHub.