package byte import ( "fmt" "os" "testing" "go.jknobloc.com/x/shelf" ) func TestNewTokenizer_Encode(t *testing.T) { var file *os.File if f, err := os.Open(shelf.Abs("tokenizers/minipile/tokenizer_gpt2_256_m000_minipile/vocab.json")); err != nil { t.Fatal(err) } else { file = f } defer file.Close() tok, err := NewTokenizer(file) if err != nil { t.Fatal(err) } fmt.Println(tok.Encode(" \nabc")) } func TestNewTokenizer_Decode(t *testing.T) { var file *os.File if f, err := os.Open(shelf.Abs("tokenizers/minipile/tokenizer_gpt2_256_m000_minipile/vocab.json")); err != nil { t.Fatal(err) } else { file = f } defer file.Close() tok, err := NewTokenizer(file) if err != nil { t.Fatal(err) } fmt.Println(tok.Decode([]uint8{220, 198, 64, 65, 66})) }