package docarray

Get desktop application:
View/edit binary Protocol Buffers messages

* Represents a (quantized) dense n-dim array

Used in: NdArrayProto, SparseNdArrayProto

• bytes buffer = 1

  the actual array data, in bytes

• repeated uint32 shape = 2

  the shape (dimensions) of the array

• string dtype = 3

  the data type of the array

Used in: jina.DataRequestProto.DataContentProto

• repeated DocumentProto docs = 1

  a list of Documents

* Represents a Document

Used in: DocumentArrayProto

• string id = 1

  A hexdigest that represents a unique document ID

• oneof content

  • bytes blob = 2

    the raw binary content of this document, which often represents the original document when comes into jina

  • NdArrayProto tensor = 3

    the ndarray of the image/audio/video document

  • string text = 4

    a text document

• uint32 granularity = 5

  the depth of the recursive chunk structure

• uint32 adjacency = 6

  the width of the recursive match structure

• string parent_id = 7

  the parent id from the previous granularity

• float weight = 8

  The weight of this document

• string uri = 9

  a uri of the document could be: a local file path, a remote url starts with http or https or data URI scheme

• string modality = 10

  modality, an identifier to the modality this document belongs to. In the scope of multi/cross modal search

• string mime_type = 11

  mime type of this document, for buffer content, this is required; for other contents, this can be guessed

• float offset = 12

  the offset of the doc

• repeated float location = 13

  the position of the doc, could be start and end index of a string; could be x,y (top, left) coordinate of an image crop; could be timestamp of an audio clip

• repeated DocumentProto chunks = 14

  list of the sub-documents of this document (recursive structure)

• repeated DocumentProto matches = 15

  the matched documents on the same level (recursive structure)

• optional NdArrayProto embedding = 16

  the embedding of this document

• optional google.protobuf.Struct tags = 17

  a structured data value, consisting of field which map to dynamically typed values.

• map<string, NamedScoreProto> scores = 18

  Scores performed on the document, each element corresponds to a metric

• map<string, NamedScoreProto> evaluations = 19

  Evaluations performed on the document, each element corresponds to a metric

• optional google.protobuf.Struct _metadata = 20

  system-defined meta attributes represented in a structured data value.

* Represents the relevance model to `ref_id`

Used in: DocumentProto

• float value = 1

  value

• string op_name = 2

  the name of the operator/score function

• string description = 3

  text description of the score

• string ref_id = 4

  the score is computed between doc `id` and `ref_id`

* Represents a general n-dim array, can be either dense or sparse

Used in: DocumentProto

• oneof content

  • DenseNdArrayProto dense = 1

    dense representation of the ndarray

  • SparseNdArrayProto sparse = 2

    sparse representation of the ndarray

• string cls_name = 3

  the name of the ndarray class

• optional google.protobuf.Struct parameters = 4

* Represents a sparse ndarray

Used in: NdArrayProto

• optional DenseNdArrayProto indices = 1

  A 2-D int64 tensor of shape [N, ndims], which specifies the indices of the elements in the sparse tensor that contain nonzero values (elements are zero-indexed)

• optional DenseNdArrayProto values = 2

  A 1-D tensor of any type and shape [N], which supplies the values for each element in indices.

• repeated uint32 shape = 3

  A 1-D int64 tensor of shape [ndims], which specifies the shape of the sparse tensor.