Array

Represents an arbitrary NumPy ndarray (1D, 2D, 3D, or higher). Supports to_numpy() for direct access. This datatype is used for generic array data transport within the Telekinesis ecosystem.

Field

Required

Field	Type	Description
value	np.ndarray	The wrapped NumPy array. Arbitrary shape and dtype (e.g. float32, int32).

Optional

None.

Methods

Method	Description
to_numpy()	Returns the underlying NumPy array.

Example

From the datatypes examples:

import numpy as np
from datatypes import datatypes
from loguru import logger

# ------------------------------------------------
# 1. Create Array instance (1D)
# ------------------------------------------------
arr_1d = np.array([1.0, 2.0, 3.0, 4.0], dtype=np.float32)
array_1d = datatypes.Array(arr_1d)

# ------------------------------------------------
# 2. Access data via to_numpy and to_list (1D case)
# ------------------------------------------------
data = array_1d.to_numpy()
logger.info("Array (1D) to_numpy shape={}", data.shape)
logger.info("Array (1D) to_numpy dtype={}", data.dtype)
logger.info("Array (1D) to_list={}", array_1d.to_list())

# ------------------------------------------------
# 3. Create Array instance (2D)
# ------------------------------------------------
arr_2d = np.array([[1, 2, 3], [4, 5, 6]], dtype=np.int32)
array_2d = datatypes.Array(arr_2d)

# ------------------------------------------------
# 4. Access data via to_numpy and to_list (2D case)
# ------------------------------------------------
numpy_arr = array_2d.to_numpy()
logger.info("Array (2D) to_numpy shape={}", numpy_arr.shape)
logger.info("Array (2D) to_numpy dtype={}", numpy_arr.dtype)
logger.info("Array (2D) to_list={}", array_2d.to_list())

# ------------------------------------------------
# 5. Create Array instance (3D)
# ------------------------------------------------
arr_3d = np.random.randn(2, 3, 4).astype(np.float32)
array_3d = datatypes.Array(arr_3d)

# ------------------------------------------------
# 6. Access data via to_numpy and to_list (3D case)
# ------------------------------------------------
data_3d = array_3d.to_numpy()
logger.info("Array (3D) to_numpy shape={}", data_3d.shape)
logger.info("Array (3D) to_numpy dtype={}", data_3d.dtype)
logger.info("Array (3D) to_list len={}", len(array_3d.to_list()))