From 6b4e704befb7f82627d2c9a4e3f4e2971fdaf883 Mon Sep 17 00:00:00 2001
From: Nick White <git@njw.name>
Date: Tue, 14 May 2019 11:20:33 +0100
Subject: Add pgconf tool that prints the overall confidence for a whole page
 of hocr

---
 pgconf/main.go | 47 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 47 insertions(+)
 create mode 100644 pgconf/main.go

(limited to 'pgconf')

diff --git a/pgconf/main.go b/pgconf/main.go
new file mode 100644
index 0000000..1b70ecc
--- /dev/null
+++ b/pgconf/main.go
@@ -0,0 +1,47 @@
+package main
+
+import (
+	"flag"
+	"fmt"
+	"log"
+	"os"
+
+	"rescribe.xyz/go.git/lib/hocr"
+	"rescribe.xyz/go.git/lib/line"
+)
+
+func main() {
+	flag.Usage = func() {
+		fmt.Fprintf(os.Stderr, "Usage: pgconf hocr\n")
+		fmt.Fprintf(os.Stderr, "Prints the total confidence for a page, as an average of the confidence of each word.\n")
+		flag.PrintDefaults()
+	}
+	flag.Parse()
+	if flag.NArg() != 1 {
+		flag.Usage()
+		os.Exit(1)
+	}
+
+	var err error
+	lines := make(line.Details, 0)
+
+	for _, f := range flag.Args() {
+		var newlines line.Details
+		newlines, err = hocr.GetLineBasics(f)
+		if err != nil {
+			log.Fatal(err)
+		}
+
+		for _, l := range newlines {
+			lines = append(lines, l)
+		}
+	}
+
+	var total float64
+	for _, l := range lines {
+		total += l.Avgconf
+	}
+	avg := total / float64(len(lines))
+
+	fmt.Printf("%0.2f\n", avg)
+}
-- 
cgit v1.2.1-24-ge1ad