Convert the input to a dask array. Subclasses of ``np.ndarray`` will be passed through as chunks unchanged. Parameters ---------- a : array-like Input data, in any form that can be converted to a dask array. This includes lists, lists of tuples, tuples, tuples of tu
(a, dtype=None, order=None, *, like=None, inline_array=False)
| 1281 | |
| 1282 | |
| 1283 | def asanyarray(a, dtype=None, order=None, *, like=None, inline_array=False): |
| 1284 | """Convert the input to a dask array. |
| 1285 | |
| 1286 | Subclasses of ``np.ndarray`` will be passed through as chunks unchanged. |
| 1287 | |
| 1288 | Parameters |
| 1289 | ---------- |
| 1290 | a : array-like |
| 1291 | Input data, in any form that can be converted to a dask array. This |
| 1292 | includes lists, lists of tuples, tuples, tuples of tuples, tuples of |
| 1293 | lists and ndarrays. |
| 1294 | dtype : data-type, optional |
| 1295 | By default, the data-type is inferred from the input data. |
| 1296 | order : {‘C’, ‘F’, ‘A’, ‘K’}, optional |
| 1297 | Memory layout. ‘A’ and ‘K’ depend on the order of input array a. |
| 1298 | ‘C’ row-major (C-style), ‘F’ column-major (Fortran-style) memory |
| 1299 | representation. ‘A’ (any) means ‘F’ if a is Fortran contiguous, ‘C’ |
| 1300 | otherwise ‘K’ (keep) preserve input order. Defaults to ‘C’. |
| 1301 | like: array-like |
| 1302 | Reference object to allow the creation of Dask arrays with chunks |
| 1303 | that are not NumPy arrays. If an array-like passed in as ``like`` |
| 1304 | supports the ``__array_function__`` protocol, the chunk type of the |
| 1305 | resulting array will be defined by it. In this case, it ensures the |
| 1306 | creation of a Dask array compatible with that passed in via this |
| 1307 | argument. If ``like`` is a Dask array, the chunk type of the |
| 1308 | resulting array will be defined by the chunk type of ``like``. |
| 1309 | Requires NumPy 1.20.0 or higher. |
| 1310 | inline_array: |
| 1311 | Whether to inline the array in the resulting dask graph. For more information, |
| 1312 | see the documentation for ``dask.array.from_array()``. |
| 1313 | |
| 1314 | Returns |
| 1315 | ------- |
| 1316 | out : dask array |
| 1317 | Dask array interpretation of a. |
| 1318 | |
| 1319 | Examples |
| 1320 | -------- |
| 1321 | >>> import dask.array as da |
| 1322 | >>> import numpy as np |
| 1323 | >>> x = np.arange(3) |
| 1324 | >>> da.asanyarray(x) |
| 1325 | dask.array<array, shape=(3,), dtype=int64, chunksize=(3,), chunktype=numpy.ndarray> |
| 1326 | |
| 1327 | >>> y = [[1, 2, 3], [4, 5, 6]] |
| 1328 | >>> da.asanyarray(y) |
| 1329 | dask.array<array, shape=(2, 3), dtype=int64, chunksize=(2, 3), chunktype=numpy.ndarray> |
| 1330 | |
| 1331 | .. warning:: |
| 1332 | `order` is ignored if `a` is an `Array`, has the attribute ``to_dask_array``, |
| 1333 | or is a list or tuple of `Array`'s. |
| 1334 | """ |
| 1335 | if like is None: |
| 1336 | if isinstance(a, Array): |
| 1337 | return _as_dtype(a, dtype) |
| 1338 | elif hasattr(a, "to_dask_array"): |
| 1339 | return _as_dtype(a.to_dask_array(), dtype) |
| 1340 | elif type(a).__module__.split(".")[0] == "xarray" and hasattr(a, "data"): |
no test coverage detected
searching dependent graphs…