embedworkflows

package
v0.8.3 Latest Latest
Warning

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

Go to latest
Published: Jun 5, 2023 License: MIT Imports: 7 Imported by: 2

Documentation

Index

Constants

This section is empty.

Variables

This section is empty.

Functions

func ConfirmEmbeddingSearch added in v0.8.0

func ConfirmEmbeddingSearch(yes bool) types.NamedWorkflow[*tzap.Tzap, *tzap.Tzap]

func FetchOrCachedEmbeddingForFilesWorkflow

func FetchOrCachedEmbeddingForFilesWorkflow() types.NamedWorkflow[*tzap.Tzap, *tzap.Tzap]

func LoadAndFetchEmbeddings added in v0.8.0

func LoadAndFetchEmbeddings(files []types.FileReader, embedder *embed.Embedder, disableIndex, yes bool) types.NamedWorkflow[*tzap.Tzap, *tzap.Tzap]

func PrepareEmbedFilesWorkflow

func PrepareEmbedFilesWorkflow(files []types.FileReader, embedder *embed.Embedder) types.NamedWorkflow[*tzap.Tzap, *tzap.Tzap]

func SaveAndLoadEmbeddingsToDB

func SaveAndLoadEmbeddingsToDB() types.NamedWorkflow[*tzap.Tzap, *tzap.Tzap]

func SearchFilesWorkflow

func SearchFilesWorkflow(query types.QueryRequest, excludeFiles []string, k int, n int) types.NamedWorkflow[*tzap.Tzap, *tzap.Tzap]

k is amount of embeddings to be included. When using inspiration files, embeddings are likely to be duplicated and as such are filtered out. n is used to increase how many embeddings are fetched but are trimmed to only contain top K after filtering.

Types

This section is empty.

Jump to

Keyboard shortcuts

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