corpus

package
v0.3.0 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Apr 2, 2026 License: MIT Imports: 6 Imported by: 0

Documentation

Index

Constants

View Source 
const (
	VectorLen = 1024

	PrefixQ = "Q:"
	PrefixA = "A:"
)
View Source 
const (
	DocumentTable = "corpus_document"
	DocumentAlias = "cd"
	DocumentLabel = "document"
	DocumentTypID = "corpusDocument"
)

consts of Document 文档

View Source 
const (
	DocVectorTable = "corpus_vector_400"
	DocVectorAlias = "cv"
	DocVectorLabel = "docVector"
	DocVectorTypID = "corpusDocVector"
)

consts of DocVector 文档向量

View Source 
const (
	DocMatchLabel = "docMatch"
	DocMatchTypID = "corpusDocMatch"
)

consts of DocMatch 提示匹配结果

View Source 
const (
	ChatLogTable = "qa_chat_log"
	ChatLogAlias = "cl"
	ChatLogLabel = "chatLog"
	ChatLogTypID = "corpusChatLog"
)

consts of ChatLog 聊天日志

Variables

This section is empty.

Functions

This section is empty.

Types

type ChatLog 

type ChatLog struct {
	comm.BaseModel `bun:"table:qa_chat_log,alias:cl" json:"-"`

	comm.DefaultModel

	ChatLogBasic

	comm.MetaField

} // @name corpusChatLog

ChatLog 聊天日志 Deprecated

func NewChatLogWithBasic 

func NewChatLogWithBasic(in ChatLogBasic) *ChatLog

func NewChatLogWithID 

func NewChatLogWithID(id any) *ChatLog

func (*ChatLog) Creating 

func (z *ChatLog) Creating() error

Creating function call to it's inner fields defined hooks

func (*ChatLog) IdentityAlias 

func (_ *ChatLog) IdentityAlias() string

func (*ChatLog) IdentityLabel 

func (_ *ChatLog) IdentityLabel() string

func (*ChatLog) IdentityModel 

func (_ *ChatLog) IdentityModel() string

func (*ChatLog) IdentityTable 

func (_ *ChatLog) IdentityTable() string

func (*ChatLog) SetWith 

func (z *ChatLog) SetWith(o ChatLogSet)

type ChatLogBasic 

type ChatLogBasic struct {
	// 会话ID
	ChatID oid.OID `bun:"csid,notnull" extensions:"x-order=A" json:"csid" pg:"csid,notnull" swaggertype:"string"`
	// 提问
	Question string `bun:",notnull,type:text" extensions:"x-order=B" form:"prompt" json:"prompt" pg:",notnull,type:text"`
	// 回答
	Answer string `bun:",notnull,type:text" extensions:"x-order=C" form:"response" json:"response" pg:",notnull,type:text"`
	// for meta update
	MetaDiff *comm.MetaDiff `bson:"-" bun:"-" json:"metaUp,omitempty" pg:"-" swaggerignore:"true"`

} // @name corpusChatLogBasic

func (*ChatLogBasic) MetaAddKVs 

func (in *ChatLogBasic) MetaAddKVs(args ...any) *ChatLogBasic

type ChatLogSet 

type ChatLogSet struct {
	// 提问
	Question *string `extensions:"x-order=A" json:"prompt"`
	// 回答
	Answer *string `extensions:"x-order=B" json:"response"`
	// for meta update
	MetaDiff *comm.MetaDiff `json:"metaUp,omitempty" swaggerignore:"true"`

} // @name corpusChatLogSet

func (*ChatLogSet) MetaAddKVs 

func (in *ChatLogSet) MetaAddKVs(args ...any) *ChatLogSet

type ChatLogs 

type ChatLogs []ChatLog

type DocMatch 

type DocMatch struct {
	// 文档编号
	DocID oid.OID `bun:"doc_id" extensions:"x-order=A" json:"docID" swaggertype:"string"`
	// 提示
	Subject string `bun:"subject" extensions:"x-order=B" form:"subject" json:"subject"`
	// 相似度
	Similarity float32 `bun:"similarity" extensions:"x-order=C" json:"similarity,omitempty"`

} // @name corpusDocMatch

DocMatch 提示匹配结果

type DocMatches 

type DocMatches []DocMatch

func (DocMatches) DocumentIDs 

func (z DocMatches) DocumentIDs() (out oid.OIDs)

DocumentIDs returns deduplicated document IDs from the document match list

func (DocMatches) Subjects 

func (z DocMatches) Subjects(n int) (out []string)

Subjects returns all document subjects from the document match list

type DocVector 

type DocVector struct {
	comm.BaseModel `bun:"table:corpus_vector_400,alias:cv" json:"-"`

	comm.DefaultModel

	DocVectorBasic

	// 相似度 仅用于查询结果
	Similarity float32 `bun:"-" extensions:"x-order=D" json:"similarity,omitempty" pg:"-"`

	comm.MetaField

} // @name corpusDocVector

DocVector 文档向量 400=1024, 600=1536

func NewDocVectorWithBasic 

func NewDocVectorWithBasic(in DocVectorBasic) *DocVector

func NewDocVectorWithID 

func NewDocVectorWithID(id any) *DocVector

func (*DocVector) Creating 

func (z *DocVector) Creating() error

Creating function call to it's inner fields defined hooks

func (*DocVector) IdentityAlias 

func (_ *DocVector) IdentityAlias() string

func (*DocVector) IdentityLabel 

func (_ *DocVector) IdentityLabel() string

func (*DocVector) IdentityModel 

func (_ *DocVector) IdentityModel() string

func (*DocVector) IdentityTable 

func (_ *DocVector) IdentityTable() string

func (*DocVector) SetWith 

func (z *DocVector) SetWith(o DocVectorSet)

type DocVectorBasic 

type DocVectorBasic struct {
	// 文档编号
	DocID oid.OID `bun:"doc_id,notnull" extensions:"x-order=A" json:"docID" pg:"doc_id,notnull" swaggertype:"string"`
	// 主题 由名称+属性组成
	Subject string `bun:"subject,notnull,type:text" extensions:"x-order=B" form:"subject" json:"subject" pg:"subject,notnull,type:text"`
	// 向量值 长为1024的浮点数集
	Vector Vector `bun:"embedding,type:vector(1024)" extensions:"x-order=C" json:"vector,omitempty" pg:"embedding,type:vector(1024)"`
	// for meta update
	MetaDiff *comm.MetaDiff `bson:"-" bun:"-" json:"metaUp,omitempty" pg:"-" swaggerignore:"true"`

} // @name corpusDocVectorBasic

func (*DocVectorBasic) MetaAddKVs 

func (in *DocVectorBasic) MetaAddKVs(args ...any) *DocVectorBasic

type DocVectorSet 

type DocVectorSet struct {
	// 主题 由名称+属性组成
	Subject *string `extensions:"x-order=A" json:"subject"`
	// 向量值 长为1024的浮点数集
	Vector *Vector `extensions:"x-order=B" json:"vector,omitempty"`
	// for meta update
	MetaDiff *comm.MetaDiff `json:"metaUp,omitempty" swaggerignore:"true"`

} // @name corpusDocVectorSet

func (*DocVectorSet) MetaAddKVs 

func (in *DocVectorSet) MetaAddKVs(args ...any) *DocVectorSet

type DocVectors 

type DocVectors []DocVector

type Document 

type Document struct {
	comm.BaseModel `bun:"table:corpus_document,alias:cd" json:"-"`

	comm.DefaultModel

	DocumentBasic

	comm.MetaField

} // @name corpusDocument

Document 文档 语料库

func NewDocumentWithBasic 

func NewDocumentWithBasic(in DocumentBasic) *Document

func NewDocumentWithID 

func NewDocumentWithID(id any) *Document

func (*Document) Creating 

func (z *Document) Creating() error

Creating function call to it's inner fields defined hooks

func (*Document) GetFullText added in v0.2.3 

func (z *Document) GetFullText() string

func (*Document) GetSubject 

func (z *Document) GetSubject() string

GetSubject returns the document subject (title + heading)

func (*Document) IdentityAlias 

func (_ *Document) IdentityAlias() string

func (*Document) IdentityLabel 

func (_ *Document) IdentityLabel() string

func (*Document) IdentityModel 

func (_ *Document) IdentityModel() string

func (*Document) IdentityTable 

func (_ *Document) IdentityTable() string

func (*Document) SetWith 

func (z *Document) SetWith(o DocumentSet)

type DocumentBasic 

type DocumentBasic struct {
	// 主标题 名称
	Title string `` /* 161-byte string literal not displayed */
	// 小节标题 属性 类别
	Heading string `` /* 165-byte string literal not displayed */
	// 内容 值
	Content string `bun:",notnull,type:text" extensions:"x-order=C" form:"content" json:"content" pg:",notnull,type:text"`
	// for meta update
	MetaDiff *comm.MetaDiff `bson:"-" bun:"-" json:"metaUp,omitempty" pg:"-" swaggerignore:"true"`

} // @name corpusDocumentBasic

func (*DocumentBasic) MetaAddKVs 

func (in *DocumentBasic) MetaAddKVs(args ...any) *DocumentBasic

type DocumentSet 

type DocumentSet struct {
	// 主标题 名称
	Title *string `extensions:"x-order=A" json:"title"`
	// 小节标题 属性 类别
	Heading *string `extensions:"x-order=B" json:"heading"`
	// 内容 值
	Content *string `extensions:"x-order=C" json:"content"`
	// for meta update
	MetaDiff *comm.MetaDiff `json:"metaUp,omitempty" swaggerignore:"true"`

} // @name corpusDocumentSet

func (*DocumentSet) MetaAddKVs 

func (in *DocumentSet) MetaAddKVs(args ...any) *DocumentSet

type Documents 

type Documents []Document

func (Documents) Headings 

func (z Documents) Headings() []string

Headings returns all document headings in the document list

func (Documents) IDs 

func (z Documents) IDs() (out oid.OIDs)

IDs returns all document IDs in the document list

func (Documents) MarkdownText 

func (z Documents) MarkdownText() string

MarkdownText converts document list to Markdown format text for LLM context

type Pair 

type Pair struct {
	Question string `json:"Q"`
	Anwser   string `json:"A"`
}

Pair 一对问答

type Pairs 

type Pairs []Pair

Pairs is a list of Q&A pairs

func ParseText 

func ParseText(text string) (result Pairs)

ParseText parses text into a list of Q&A pairs, supports Q: and A: prefix format

type Vector 

type Vector []float32

Vector is the vector type for document embedding

Source Files

View all Source files

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.