summaryrefslogtreecommitdiff
path: root/dataset/cmd
diff options
context:
space:
mode:
Diffstat (limited to 'dataset/cmd')
-rw-r--r--dataset/cmd/dataset/main.go43
1 files changed, 43 insertions, 0 deletions
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()
+}