inference

package

v0.0.0-...-7b461c4 Latest Latest Go to latest Published: Mar 12, 2024 License: Apache-2.0 Imports: 12 Imported by: 0

Details

Valid go.mod file

The Go module system was introduced in Go 1.11 and is the official dependency management solution for Go.
Redistributable license

Redistributable licenses place minimal restrictions on how software can be used, modified, and redistributed.
Tagged version

Modules with tagged versions give importers more predictable builds.
Stable version

When a project reaches major version v1 it is considered stable.
Learn more about best practices

Repository

github.com/adalkiran/llama-nuts-and-bolts

Documentation ¶

Index ¶

type GeneratedPart
type GenerationState
type InferenceEngine
- func NewInferenceEngine(model *model.Model, inferenceArgs common.InferenceArgs, ...) *InferenceEngine
type RuneData
type TokenGeneratorFn

Constants ¶

This section is empty.

Variables ¶

This section is empty.

Functions ¶

This section is empty.

Types ¶

type GeneratedPart ¶

type GeneratedPart struct {
	DecodedString        string
	TokenId              model.TokenId
	Token                sentencepiece.SentencePiece
	AddedToWaiting       bool
	WaitingRunesExtraStr string
	IsResendOfWaiting    bool
	GenerationState      GenerationState
}

type GenerationState ¶

type GenerationState byte

const (
	GSInProgress               GenerationState = 1
	GSFinishedByReachingEOS    GenerationState = 2
	GSFinishedByReachingSeqLen GenerationState = 3
)

type InferenceEngine ¶

type InferenceEngine struct {
	// contains filtered or unexported fields
}

func NewInferenceEngine ¶

func NewInferenceEngine(model *model.Model, inferenceArgs common.InferenceArgs, logFn func(format string, v ...any)) *InferenceEngine

func (*InferenceEngine) CreateInferenceContext ¶

func (ie *InferenceEngine) CreateInferenceContext() *model.InferenceContext

func (*InferenceEngine) GenerateString ¶

func (ie *InferenceEngine) GenerateString(promptTokens []model.TokenId) (<-chan GeneratedPart, <-chan error)

func (*InferenceEngine) GenerateStringFromOutputTokens ¶

func (ie *InferenceEngine) GenerateStringFromOutputTokens(outputTokens []model.TokenId) (<-chan GeneratedPart, <-chan error)

func (*InferenceEngine) GenerateStringGeneric ¶

func (ie *InferenceEngine) GenerateStringGeneric(promptTokens []model.TokenId, tokenGeneratorFn TokenGeneratorFn) (<-chan GeneratedPart, <-chan error)

func (*InferenceEngine) GenerateTokensGeneric ¶

func (ie *InferenceEngine) GenerateTokensGeneric(promptTokens []model.TokenId, tokenGeneratorFn TokenGeneratorFn) (<-chan generationStepResult[model.TokenId], <-chan error)

func (*InferenceEngine) TokenBatchToDebugString ¶

func (ie *InferenceEngine) TokenBatchToDebugString(tokenIdBatch []model.TokenId) string

func (*InferenceEngine) TokenBatchToString ¶

func (ie *InferenceEngine) TokenBatchToString(tokenIdBatch []model.TokenId) ([]sentencepiece.SentencePiece, string)

func (*InferenceEngine) TokenToString ¶

func (ie *InferenceEngine) TokenToString(tokenId model.TokenId, decodingContext *generationDecodingContext) (token sentencepiece.SentencePiece, resultString string, addedToWaiting bool)

func (*InferenceEngine) Tokenize ¶

func (ie *InferenceEngine) Tokenize(text string, addBeginOfSentence bool) ([]model.TokenId, error)

func (*InferenceEngine) TokenizeBatch ¶

func (ie *InferenceEngine) TokenizeBatch(texts []string, addBeginOfSentence bool) (result [][]model.TokenId, err error)

type RuneData ¶

type RuneData struct {
	Rune          rune
	RuneName      string
	UnicodeEscape string
}

type TokenGeneratorFn ¶

type TokenGeneratorFn = func(promptTokens []model.TokenId, generatedTokensCh chan<- generationStepResult[model.TokenId], errorCh chan<- error)

Source Files ¶

View all Source files

?	: This menu
/	: Search site
f or F	: Jump to
y or Y	: Canonical URL