pyflink.multimodal.expression.ImageExpressionAccessor.to_tensor

ImageExpressionAccessor.to_tensor(*, width: int, height: int, mode: str = 'RGB', layout: Literal['CHW', 'HWC'] = 'CHW', concurrency: Optional[int] = None) → pyflink.table.expression.Expression

Convert already prepared image values to fixed-shape tensors.

Equivalent to image_to_tensor(). See that function for full parameter details. This operation does not resize or convert image mode. Use .image.resize(...) or .image.convert_mode(...) upstream when needed.

Parameters

• width – Expected input image width in pixels.

• height – Expected input image height in pixels.

• mode – Expected input image mode before tensor conversion.

• layout – Tensor layout, such as "CHW" or "HWC".

• concurrency – Optional execution concurrency for this operation. None uses the framework default.

Returns

A DataFrame expression that produces a FLOAT32 tensor with the requested shape and layout.