schema.common.rnc Maven / Gradle / Ivy

# Copyright Vespa.ai. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root.
service.attlist &= attribute hostalias { xsd:NCName }
service.attlist &= attribute baseport { xsd:unsignedShort }?
service.attlist &= attribute jvm-options { text }? # Remove in Vespa 9
service.attlist &= attribute jvm-gc-options { text }? # Remove in Vespa 9
# preload is for internal use only
service.attlist &= attribute preload { text }?

anyElement = element * {
   (attribute * { text }
    | text
    | anyElement)*
}

# Valid qualified java class name. See http://docs.oracle.com/javase/specs/jls/se8/html/jls-3.html#jls-3.8
JavaId = xsd:string { pattern = "([a-zA-Z_$][a-zA-Z\d_$]*\.)*[a-zA-Z_$][a-zA-Z\d_$]*" }

Nodes = element nodes {
    attribute count { xsd:positiveInteger | xsd:string } &
    attribute flavor { xsd:string }? &
    attribute docker-image { xsd:string }? &
    Resources?
}

Resources = element resources {
    attribute vcpu { xsd:double { minExclusive = "0.0" } | xsd:string }? &
    attribute memory { xsd:string }? &
    attribute disk { xsd:string }? &
    attribute disk-speed { xsd:string }? &
    attribute storage-type { xsd:string }? &
    attribute architecture { xsd:string }? &
    GpuResources?
}

GpuResources = element gpu {
  attribute count { xsd:positiveInteger } &
  attribute memory { xsd:string }
}

OptionalDedicatedNodes = element nodes {
    attribute count { xsd:positiveInteger | xsd:string } &
    attribute flavor { xsd:string }? &
    attribute required { xsd:boolean }? &
    attribute docker-image { xsd:string }? &
    attribute dedicated { xsd:boolean }? &
    attribute exclusive { xsd:boolean }? &
    Resources?
}

GenericConfig = element config {
    attribute name { text },
    attribute version { text }?,
    anyElement*
}

ModelReference =
    attribute model-id { xsd:string }? &
    attribute path { xsd:string }? &
    attribute url { xsd:string }?

ComponentSpec =
    ( attribute id { xsd:Name | JavaId } | attribute idref { xsd:Name } | attribute ident { xsd:Name } )

ComponentId =
    ComponentSpec

BundleSpec =
    attribute class { xsd:Name | JavaId }? &
    attribute bundle { xsd:Name }?

Component = element component {
   (ComponentDefinition | TypedComponentDefinition)
}

ComponentDefinition =
   ComponentId &
   BundleSpec &
   GenericConfig* &
   Component*

TypedComponentDefinition =
   attribute id { xsd:Name } &
   (HuggingFaceEmbedder | HuggingFaceTokenizer | BertBaseEmbedder | ColBertEmbedder | SpladeEmbedder ) &
   GenericConfig* &
   Component*

HuggingFaceEmbedder =
   attribute type { "hugging-face-embedder" } &
   element transformer-model { ModelReference } &
   element tokenizer-model { ModelReference }? &
   element max-tokens { xsd:positiveInteger }? &
   element transformer-input-ids { xsd:string }? &
   element transformer-attention-mask { xsd:string }? &
   element transformer-token-type-ids { xsd:string }? &
   element transformer-output { xsd:string }? &
   element normalize { xsd:boolean }? &
   PrependResources? &
   OnnxModelExecutionParams &
   EmbedderPoolingStrategy

PrependResources = element prepend {
  element query { xsd:string }? &
  element document { xsd:string }?
}

SpladeEmbedder =
   attribute type { "splade-embedder" } &
   element transformer-model { ModelReference } &
   element tokenizer-model { ModelReference }? &
   element max-tokens { xsd:positiveInteger }? &
   element transformer-input-ids { xsd:string }? &
   element transformer-attention-mask { xsd:string }? &
   element transformer-token-type-ids { xsd:string }? &
   element transformer-output { xsd:string }? &
   element term-score-threshold { xsd:double }? &
   OnnxModelExecutionParams

HuggingFaceTokenizer =
    attribute type { "hugging-face-tokenizer" } &
    element model { attribute language { xsd:string }? & ModelReference }+

BertBaseEmbedder =
    attribute type { "bert-embedder" } &
    element transformer-model { ModelReference } &
    element tokenizer-vocab { ModelReference } &
    element max-tokens { xsd:nonNegativeInteger }? &
    element transformer-input-ids { xsd:string }? &
    element transformer-attention-mask { xsd:string }? &
    element transformer-token-type-ids { xsd:string }? &
    element transformer-output { xsd:string }? &
    StartOfSequence &
    EndOfSequence &
    OnnxModelExecutionParams &
    EmbedderPoolingStrategy


ColBertEmbedder =
   attribute type { "colbert-embedder" } &
   element transformer-model { ModelReference } &
   element tokenizer-model { ModelReference }? &
   element max-tokens { xsd:positiveInteger }? &
   element max-query-tokens { xsd:positiveInteger }? &
   element max-document-tokens { xsd:positiveInteger }? &
   element transformer-mask-token { xsd:integer }? &
   element transformer-input-ids { xsd:string }? &
   element transformer-attention-mask { xsd:string }? &
   element transformer-pad-token { xsd:integer }? &
   element query-token-id { xsd:integer }? &
   element document-token-id { xsd:integer }? &
   element transformer-output { xsd:string }? &
   OnnxModelExecutionParams &
   StartOfSequence &
   EndOfSequence

OnnxModelExecutionParams =
    element onnx-execution-mode { "parallel" | "sequential" }? &
    element onnx-interop-threads { xsd:integer }? &
    element onnx-intraop-threads { xsd:integer }? &
    element onnx-gpu-device { xsd:integer }?

EmbedderPoolingStrategy = element pooling-strategy { "cls" | "mean" }?

StartOfSequence = element transformer-start-sequence-token { xsd:integer }?
EndOfSequence = element transformer-end-sequence-token { xsd:integer }?