API reference¶

Core¶

classtensorstore.TensorStore

Asynchronous multi-dimensional array handle.

classtensorstore.Context

Manages shared TensorStorecontext resources, such as caches and credentials.

classtensorstore.KvStore

Key-value store that maps an ordered set of byte string keys to byte string values.

classtensorstore.Transaction

Transactions are used to stage a group of modifications (e.g. writes totensorstore.TensorStore objects) in memory, and then either commit thegroup all at once or abort it.

classtensorstore.Batch

Batches are used to group together read operations for potentially improvedefficiency.

tensorstore.open(spec:Spec|Any,*,...)→Future[TensorStore]

Opens or creates aTensorStore from aSpec.

Indexing¶

tensorstore.inf:int=4611686018427387903

Special constant equal to\(2^{62}-1\) that indicates an unboundedindex domain.

classtensorstore.IndexDomain

Domain (including bounds and optional dimension labels) of an N-dimensionalindex space.

classtensorstore.IndexTransform

Represents a transform from an input index space to an output space.

classtensorstore.Dim

1-d index interval with optionally-implicit bounds and dimension label.

classtensorstore.OutputIndexMap

Represents an output index map for an index transform.

classtensorstore.OutputIndexMaps

View of the output index maps for an index transform.

classtensorstore.OutputIndexMethod

Indicates theoutput index method of anOutputIndexMap.

classtensorstore.DimExpression

Specifies an advanced indexing operation.

classtensorstore.DimSelection(DimExpression)

Specifies a dimension selection, for starting adimension expression.

statictensorstore.d[selection:DimSelectionLike]→DimSelection

Constructs aDimSelection from a sequence of dimension indices, ranges, and/or labels.

tensorstore.newaxis=None

Alias forNone used inindexing expressions to specify a new singleton dimension.

classtensorstore.Indexable

Abstract base class for types that supportTensorStore indexing operations.

tensorstore.NumpyIndexTerm

Individual term in aNumpyIndexingSpec.

tensorstore.NumpyIndexingSpec

NumPy-style indexing expression.

tensorstore.DimSelectionLike

Subsequence of dimensions to select in aDimSelection object.

Spec¶

classtensorstore.Spec

Specification for opening or creating aTensorStore.

classtensorstore.Schema

Driver-independent options for defining a TensorStore schema.

classtensorstore.CodecSpec

Specifies driver-specific encoding/decoding parameters.

classtensorstore.OpenMode

Specifies the mode to use when opening aTensorStore.

classtensorstore.ChunkLayout

Describes the storage layout of atensorstore.TensorStore.

classtensorstore.Unit

Specifies a physical quantity/unit.

tensorstore.RecheckCacheOption

Determines under what circumstances cached data is revalidated.

Views¶

tensorstore.DownsampleMethod

Downsampling method for use by thedownsample driver.

tensorstore.cast(base:TensorStore,dtype:DTypeLike)→TensorStore

Returns a read/write view with the data type converted.

tensorstore.cast(base:Spec,dtype:DTypeLike)→Spec

Returns a view with the data type converted.

tensorstore.array(array:ArrayLike,...)→TensorStore

Returns a TensorStore that reads/writes from an in-memory array.

tensorstore.overlay(layers,...)→TensorStore

Virtually overlays a sequence ofTensorStore layers within a common domain.

tensorstore.stack(layers,...)→TensorStore

Virtually stacks a sequence ofTensorStore layers along a new dimension.

tensorstore.concat(layers,...)→TensorStore

Virtually concatenates a sequence ofTensorStore layers along an existing dimension.

tensorstore.downsample(base:TensorStore,...)→TensorStore

Returns a virtualdownsampled view of aTensorStore.

tensorstore.downsample(base:Spec,downsample_factors,...)→Spec

Returns a virtualdownsampled view view of aSpec.

Virtual views¶

classtensorstore.VirtualChunkedReadParameters

Options passed to read callbacks used withvirtual_chunked.

classtensorstore.VirtualChunkedWriteParameters

Options passed to write callbacks used withvirtual_chunked.

tensorstore.virtual_chunked(...)→TensorStore

Creates aTensorStore where the content is read/written chunk-wise by an arbitrary function.

Data types¶

classtensorstore.dtype

TensorStore data type representation.

tensorstore.bool:dtype=dtype("bool")

Boolean data type (0 or 1). Corresponds to thebool type andnumpy.bool_.

tensorstore.char:dtype=dtype("char")

Single byte, interpreted as an ASCII character.

tensorstore.byte:dtype=dtype("byte")

Single byte.

tensorstore.int2:dtype=dtype("int2")

2-bit signedtwo’s-complement integer data type, internally stored as its 8-bit signed integer equivalent (i.e. sign-extended). Corresponds tojax.numpy.int2.

tensorstore.int4:dtype=dtype("int4")

4-bit signedtwo’s-complement integer data type, internally stored as its 8-bit signed integer equivalent (i.e. sign-extended). Corresponds tojax.numpy.int4.

tensorstore.int8:dtype=dtype("int8")

8-bit signedtwo’s-complement integer data type. Corresponds tonumpy.int8.

tensorstore.uint8:dtype=dtype("uint8")

8-bit unsigned integer. Corresponds tonumpy.uint8.

tensorstore.int16:dtype=dtype("int16")

16-bit signedtwo’s-complement integer data type. Corresponds tonumpy.int16.

tensorstore.uint16:dtype=dtype("uint16")

16-bit unsigned integer. Corresponds tonumpy.uint16.

tensorstore.int32:dtype=dtype("int32")

32-bit signedtwo’s-complement integer data type. Corresponds tonumpy.int32.

tensorstore.uint32:dtype=dtype("uint32")

32-bit unsigned integer. Corresponds tonumpy.uint32.

tensorstore.int64:dtype=dtype("int64")

32-bit signedtwo’s-complement integer data type. Corresponds tonumpy.int64.

tensorstore.uint64:dtype=dtype("uint64")

64-bit unsigned integer data type. Corresponds tonumpy.uint64.

tensorstore.float8_e3m4:dtype=dtype("float8_e3m4")

8-bit floating-point data type.

tensorstore.float8_e4m3fn:dtype=dtype("float8_e4m3fn")

8-bit floating-point data type.

tensorstore.float8_e4m3fnuz:dtype=dtype("float8_e4m3fnuz")

8-bit floating-point data type.

tensorstore.float8_e4m3b11fnuz:dtype=dtype("float8_e4m3b11fnuz")

8-bit floating-point data type.

tensorstore.float8_e5m2:dtype=dtype("float8_e5m2")

8-bit floating-point data type.

tensorstore.float8_e5m2fnuz:dtype=dtype("float8_e5m2fnuz")

8-bit floating-point data type.

tensorstore.float16:dtype=dtype("float16")

IEEE 754 binary16 half-precision floating-point data type. Correspond tonumpy.float16.

tensorstore.bfloat16:dtype=dtype("bfloat16")

bfloat16 floating-point data type.

tensorstore.float32:dtype=dtype("float32")

IEEE 754 binary32 single-precision floating-point data type. Corresponds tonumpy.float32.

tensorstore.float64:dtype=dtype("float64")

IEEE 754 binary64 double-precision floating-point data type. Corresponds tonumpy.float64.

tensorstore.complex64:dtype=dtype("complex64")

Complex number based onfloat32. Corresponds tonumpy.complex64.

tensorstore.complex128:dtype=dtype("complex128")

Complex number based onfloat64. Corresponds tonumpy.complex128.

tensorstore.string:dtype=dtype("string")

Variable-length byte string data type. Corresponds to the Pythonbytes type.

tensorstore.ustring:dtype=dtype("ustring")

Variable-length Unicode string data type. Corresponds to the Pythonstr type.

tensorstore.json:dtype=dtype("json")

JSON data type. Corresponds to an arbitrary Python JSON value.

tensorstore.DTypeLike

Value that may be converted to a TensorStoredtype.

Asynchronous support¶

classtensorstore.Future[T]

Handle forconsuming the result of an asynchronous operation.

classtensorstore.Promise[T]

Handle forproducing the result of an asynchronous operation.

classtensorstore.WriteFutures

Handle for consuming the result of an asynchronous write operation.

tensorstore.FutureLike

Generic type representing a possibly-asynchronous result.

OCDBT¶

classtensorstore.ocdbt.DistributedCoordinatorServer

Distributed coordinator server for the OCDBT (Optionally-Cooperative DistributedB+Tree) database.

tensorstore.ocdbt.dump(base:KvStore,...)→Future[Any]

Dumps the internal representation of an OCDBT database.

Experimental¶

tensorstore.experimental_collect_matching_metrics(...)→list[Any]

Collects metrics with a matching prefix.

tensorstore.experimental_collect_prometheus_format_metrics(...)→list[str]

Collects metrics in prometheus exposition format.See:https://prometheus.io/docs/instrumenting/exposition_formats/

tensorstore.experimental_push_metrics_to_prometheus(...)→Future[int]

Publishes metrics to the prometheus pushgateway.See:https://github.com/prometheus/pushgateway

tensorstore.experimental_update_verbose_logging(...)→None

Updates verbose logging flags associated with –tensorstore_verbose_logging andTENSORSTORE_VERBOSE_LOGGING flags.

tensorstore.parse_tensorstore_flags(argv:list[str])→None

Parses and initializes internal tensorstore flags from argv.