Files
llmgw/internal/control/train/pairs_test.go
T
Ilker ManapandClaude Opus 5.5 a53b2f4be9 Customise a model from the portal: a file, a model, a button
A page for anyone in a project, built for the one who knows least: an
Excel, CSV or JSONL file of questions and answers, previewed; a model; a
name; who may use it; and a button. The weights to train on are found
from where the model's came (a GGUF repository's base model), the card's
memory from their size, the node as the one with the most free. When it
is done the adapter is published by itself (a llama.cpp model's replicas
replaced to load it) and is a model of its own in the chat and the API.

The project's by default; its members share it wider (confirmed, and
never wider than the knowledge bases its pairs came from) or delete it.
One trains at a time per project. ctl train start needs only --model and
--data now, and --publish publishes when done.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-19 11:31:12 +01:00

66 lines
2.0 KiB
Go
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
package train
import (
"strings"
"testing"
)
func TestPairsFromATableByItsHeaderOrItsFirstColumns(t *testing.T) {
for _, c := range []struct {
name, file string
want []Pair
skipped int
cols [2]string
}{
{"headed.csv", "No;Cevap;Soru\n1;Yirmi gün.;İzin kaç gün?\n2;;Boş cevap\n3;Evet;Uzaktan çalışabilir miyim?\n",
[]Pair{{"İzin kaç gün?", "Yirmi gün."}, {"Uzaktan çalışabilir miyim?", "Evet"}}, 1, [2]string{"Soru", "Cevap"}},
{"plain.csv", "\ufeffİzin kaç gün?,\"Yirmi gün, yılda.\"\n\nMasraf?,Formla.\n",
[]Pair{{"İzin kaç gün?", "Yirmi gün, yılda."}, {"Masraf?", "Formla."}}, 0, [2]string{}},
{"tabs.tsv", "Question\tAnswer\nA?\tB.\n", []Pair{{"A?", "B."}}, 0, [2]string{"Question", "Answer"}},
{"d.jsonl", `{"q":"A?","a":"B."}` + "\n" + `{"question":"C?","answer":"D."}` + "\n" + `{"q":"E?"}` + "\n",
[]Pair{{"A?", "B."}, {"C?", "D."}}, 1, [2]string{}},
} {
got, err := ReadPairs(c.name, []byte(c.file))
if err != nil {
t.Errorf("%s: %v", c.name, err)
continue
}
if len(got.Pairs) != len(c.want) || got.Skipped != c.skipped || got.Columns != c.cols {
t.Errorf("%s: %+v", c.name, got)
continue
}
for i := range c.want {
if got.Pairs[i] != c.want[i] {
t.Errorf("%s: pair %d = %+v, want %+v", c.name, i, got.Pairs[i], c.want[i])
}
}
}
}
func TestPairsRefusedWithAReason(t *testing.T) {
for name, file := range map[string]string{
"a.csv": "only one column\n",
"a.xls": "x",
"a.pdf": "x",
"a.json": "not json\n",
} {
if _, err := ReadPairs(name, []byte(file)); err == nil {
t.Errorf("%s read", name)
}
}
var b strings.Builder
for range MaxPairs + 1 {
b.WriteString("x,y\n")
}
if _, err := ReadPairs("big.csv", []byte(b.String())); err == nil || !strings.Contains(err.Error(), "more than") {
t.Errorf("too many: %v", err)
}
}
func TestNDJSONIsWhatAJobReads(t *testing.T) {
got := NDJSON([]Pair{{"<b>?", "a & b"}})
if got != `{"q":"<b>?","a":"a & b"}`+"\n" {
t.Errorf("%q", got)
}
}