From 56a5c2d209192901bde81a51128c79f222f5c5be Mon Sep 17 00:00:00 2001 From: Jonas Knobloch Date: Sat, 28 Feb 2026 18:11:20 +0100 Subject: Add dataset package --- dataset/cmd/dataset/main.go | 43 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 43 insertions(+) create mode 100644 dataset/cmd/dataset/main.go (limited to 'dataset/cmd') diff --git a/dataset/cmd/dataset/main.go b/dataset/cmd/dataset/main.go new file mode 100644 index 0000000..76028be --- /dev/null +++ b/dataset/cmd/dataset/main.go @@ -0,0 +1,43 @@ +package main + +import ( + "fmt" + "log" + "path/filepath" + + "github.com/jonasknobloch/x/dataset" +) + +func main() { + root := "./tmp/minipile" + + if err := dataset.Download("JeanKaddour/minipile", "", root); err != nil { + log.Fatal(err) + } + + fmt.Println() + + split := filepath.Join(root, "train") + + var reader *dataset.Reader + + if r, err := dataset.NewReader(split); err != nil { + log.Fatal(err) + } else { + reader = r + } + + i := 0 + + for s := range reader.Texts("text") { + _ = s + fmt.Printf("\r%d", i+1) + i++ + } + + if err := reader.Err(); err != nil { + log.Fatal(err) + } + + fmt.Println() +} -- cgit v1.3.1