package bentoml.grpc.v1alpha1

Get desktop application:
View/edit binary Protocol Buffers messages

a gRPC BentoServer.

• rpc Call (Request, Response)

  service.proto:20

  Call handles methodcaller of given API entrypoint.

  message Request

  service.proto:24

  Request message for incoming Call.

  • string api_name = 1

    api_name defines the API entrypoint to call. api_name is the name of the function defined in bentoml.Service. Example: @svc.api(input=NumpyNdarray(), output=File()) def predict(input: NDArray[float]) -> bytes: ... api_name is "predict" in this case.

  • oneof content

    • NDArray ndarray = 3

      NDArray represents a n-dimensional array of arbitrary type.

    • DataFrame dataframe = 5

      DataFrame represents any tabular data type. We are using DataFrame as a trivial representation for tabular type.

    • Series series = 6

      Series portrays a series of values. This can be used for representing Series types in tabular data.

    • File file = 7

      File represents for any arbitrary file type. This can be plaintext, image, video, audio, etc.

    • google.protobuf.StringValue text = 8

      Text represents a string inputs.

    • google.protobuf.Value json = 9

      JSON is represented by using google.protobuf.Value. see https://github.com/protocolbuffers/protobuf/blob/main/src/google/protobuf/struct.proto

    • Multipart multipart = 10

      Multipart represents a multipart message. It comprises of a mapping from given type name to a subset of aforementioned types.

    • bytes serialized_bytes = 2

      serialized_bytes is for data serialized in BentoML's internal serialization format.

  message Response

  service.proto:74

  Request message for incoming Call.

  • oneof content

    • NDArray ndarray = 1

      NDArray represents a n-dimensional array of arbitrary type.

    • DataFrame dataframe = 3

      DataFrame represents any tabular data type. We are using DataFrame as a trivial representation for tabular type.

    • Series series = 5

      Series portrays a series of values. This can be used for representing Series types in tabular data.

    • File file = 6

      File represents for any arbitrary file type. This can be plaintext, image, video, audio, etc.

    • google.protobuf.StringValue text = 7

      Text represents a string inputs.

    • google.protobuf.Value json = 8

      JSON is represented by using google.protobuf.Value. see https://github.com/protocolbuffers/protobuf/blob/main/src/google/protobuf/struct.proto

    • Multipart multipart = 9

      Multipart represents a multipart message. It comprises of a mapping from given type name to a subset of aforementioned types.

    • bytes serialized_bytes = 2

      serialized_bytes is for data serialized in BentoML's internal serialization format.

DataFrame represents any tabular data type. We are using DataFrame as a trivial representation for tabular type. This message carries given implementation of tabular data based on given orientation. TODO: support index, records, etc.

Used in: Part, Request, Response

• repeated string column_names = 1

  columns name

• repeated Series columns = 2

  columns orient. { column ↠ { index ↠ value } }

File represents for any arbitrary file type. This can be plaintext, image, video, audio, etc.

Used in: Part, Request, Response

• optional File.FileType kind = 1

  optional type of file, let it be csv, text, parquet, etc.

• bytes content = 2

  contents of file as bytes.

FileType represents possible file type to be handled by BentoML. Currently, we only support plaintext (Text()), image (Image()), and file (File()). TODO: support audio and video streaming file types.

Used in: File

• FILE_TYPE_UNSPECIFIED = 0

• FILE_TYPE_CSV = 1

  file types

• FILE_TYPE_PLAINTEXT = 2

• FILE_TYPE_JSON = 3

• FILE_TYPE_BYTES = 4

• FILE_TYPE_PDF = 5

• FILE_TYPE_PNG = 6

  image types

• FILE_TYPE_JPEG = 7

• FILE_TYPE_GIF = 8

• FILE_TYPE_BMP = 9

• FILE_TYPE_TIFF = 10

• FILE_TYPE_WEBP = 11

• FILE_TYPE_SVG = 12

Multipart represents a multipart message. It comprises of a mapping from given type name to a subset of aforementioned types.

Used in: Request, Response

• map<string, Part> fields = 1

NDArray represents a n-dimensional array of arbitrary type.

Used in: Part, Request, Response

• NDArray.DType dtype = 1

  DTYPE is the data type of given array

• repeated int32 shape = 2

  shape is the shape of given array.

• repeated string string_values = 5

  represents a string parameter value.

• repeated float float_values = 3

  represents a float parameter value.

• repeated double double_values = 4

  represents a double parameter value.

• repeated bool bool_values = 6

  represents a bool parameter value.

• repeated int32 int32_values = 7

  represents a int32 parameter value.

• repeated int64 int64_values = 8

  represents a int64 parameter value.

• repeated uint32 uint32_values = 9

  represents a uint32 parameter value.

• repeated uint64 uint64_values = 10

  represents a uint64 parameter value.

Represents data type of a given array.

Used in: NDArray

• DTYPE_UNSPECIFIED = 0

  Represents a None type.

• DTYPE_FLOAT = 1

  Represents an float type.

• DTYPE_DOUBLE = 2

  Represents an double type.

• DTYPE_BOOL = 3

  Represents a bool type.

• DTYPE_INT32 = 4

  Represents an int32 type.

• DTYPE_INT64 = 5

  Represents an int64 type.

• DTYPE_UINT32 = 6

  Represents a uint32 type.

• DTYPE_UINT64 = 7

  Represents a uint64 type.

• DTYPE_STRING = 8

  Represents a string type.

Part represents possible value types for multipart message. These are the same as the types in Request message.

Used in: Multipart

• oneof representation

  • NDArray ndarray = 1

    NDArray represents a n-dimensional array of arbitrary type.

  • DataFrame dataframe = 3

    DataFrame represents any tabular data type. We are using DataFrame as a trivial representation for tabular type.

  • Series series = 5

    Series portrays a series of values. This can be used for representing Series types in tabular data.

  • File file = 6

    File represents for any arbitrary file type. This can be plaintext, image, video, audio, etc.

  • google.protobuf.StringValue text = 7

    Text represents a string inputs.

  • google.protobuf.Value json = 8

    JSON is represented by using google.protobuf.Value. see https://github.com/protocolbuffers/protobuf/blob/main/src/google/protobuf/struct.proto

  • bytes serialized_bytes = 4

    serialized_bytes is for data serialized in BentoML's internal serialization format.

Series portrays a series of values. This can be used for representing Series types in tabular data.

Used in: DataFrame, Part, Request, Response

• repeated bool bool_values = 1

  A bool parameter value

• repeated float float_values = 2

  A float parameter value

• repeated int32 int32_values = 3

  A int32 parameter value

• repeated int64 int64_values = 6

  A int64 parameter value

• repeated string string_values = 5

  A string parameter value

• repeated double double_values = 4

  represents a double parameter value.