Skip to content
Closed
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
35 changes: 23 additions & 12 deletions embedder/llamacpp/llamacpp.go
Original file line number Diff line number Diff line change
Expand Up @@ -2,10 +2,10 @@ package llamacppembedder

import (
"context"
"encoding/json"
"errors"
"os"
"os/exec"
"strconv"
"strings"

"github.com/henomis/lingoose/embedder"
)
Expand All @@ -16,6 +16,16 @@ type LlamaCppEmbedder struct {
modelPath string
}

type output struct {
Object string `json:"object"`
Data []data `json:"data"`
}
type data struct {
Object string `json:"object"`
Index int `json:"index"`
Embedding []float64 `json:"embedding"`
}

func New() *LlamaCppEmbedder {
return &LlamaCppEmbedder{
llamacppPath: "./llama.cpp/embedding",
Expand Down Expand Up @@ -61,7 +71,7 @@ func (l *LlamaCppEmbedder) embed(ctx context.Context, text string) (embedder.Emb
return nil, err
}

llamacppArgs := []string{"-m", l.modelPath, "-p", text}
llamacppArgs := []string{"-m", l.modelPath, "--embd-output-format", "json", "-p", text}
llamacppArgs = append(llamacppArgs, l.llamacppArgs...)

//nolint:gosec
Expand All @@ -74,14 +84,15 @@ func (l *LlamaCppEmbedder) embed(ctx context.Context, text string) (embedder.Emb
}

func parseEmbeddings(str string) (embedder.Embedding, error) {
strSlice := strings.Split(strings.TrimSpace(str), " ")
floatSlice := make([]float64, len(strSlice))
for i, s := range strSlice {
f, err := strconv.ParseFloat(s, 64)
if err != nil {
return nil, err
}
floatSlice[i] = f
var out output
err := json.Unmarshal([]byte(str), &out)
if err != nil {
return nil, err
}
return floatSlice, nil

if len(out.Data) != 1 {
return nil, errors.New("no embeddings found")
}

return out.Data[0].Embedding, nil
}
Loading