Add new option -l for combine_tessdata to list the network string

Signed-off-by: Stefan Weil <[email protected]>
tesseract-ocr · Jan 15, 2021 · 3195c8f · 3195c8f
1 parent 970eba7
commit 3195c8f
Show file tree

Hide file tree

Showing 3 changed files with 30 additions and 2 deletions.
diff --git a/doc/combine_tessdata.1.asc b/doc/combine_tessdata.1.asc
@@ -66,6 +66,9 @@ OPTIONS
 *-e* '.traineddata' 'FILE'...:
     Extracts the specified components from the .traineddata file
 
+*-l* '.traineddata' 'FILE'...:
+   List the network information.
+
 *-o* '.traineddata' 'FILE'...:
     Overwrites the specified components of the .traineddata file
     with those provided on the command line.

diff --git a/src/lstm/lstmrecognizer.h b/src/lstm/lstmrecognizer.h
@@ -101,6 +101,12 @@ class TESS_API LSTMRecognizer {
       return learning_rate_;
     }
   }
+
+  // Return the network string.
+  const char* GetNetwork() const {
+    return network_str_.c_str();
+  }
+
   // Multiplies the all the learning rate(s) by the given factor.
   void ScaleLearningRate(double factor) {
     ASSERT_HOST(network_ != nullptr && network_->type() == NT_SERIES);

diff --git a/src/training/combine_tessdata.cpp b/src/training/combine_tessdata.cpp
@@ -22,6 +22,7 @@
 #include "tessdatamanager.h"
 
 #include <cerrno>
+#include <iostream>             // std::cout
 
 using namespace tesseract;
 
@@ -174,6 +175,21 @@ int main(int argc, char **argv) {
   } else if (argc == 3 && strcmp(argv[1], "-d") == 0) {
     // Initialize TessdataManager with the data in the given traineddata file.
     tm.Init(argv[2]);
+  } else if (argc == 3 && strcmp(argv[1], "-l") == 0) {
+    if (!tm.Init(argv[2])) {
+      tprintf("Failed to read %s\n", argv[2]);
+      return EXIT_FAILURE;
+    }
+    tesseract::TFile fp;
+    if (tm.GetComponent(tesseract::TESSDATA_LSTM, &fp)) {
+      tesseract::LSTMRecognizer recognizer;
+      if (!recognizer.DeSerialize(&tm, &fp)) {
+        tprintf("Failed to deserialize LSTM in %s!\n", argv[2]);
+        return EXIT_FAILURE;
+      }
+      std::cout << recognizer.GetNetwork() << '\n';
+    }
+    return EXIT_SUCCESS;
   } else {
     printf("Usage for combining tessdata components:\n"
            "  %s language_data_path_prefix\n"
@@ -188,10 +204,13 @@ int main(int argc, char **argv) {
            argv[0], argv[0]);
     printf("Usage for unpacking all tessdata components:\n"
            "  %s -u traineddata_file output_path_prefix\n"
-           "  (e.g. %s -u eng.traineddata tmp/eng.)\n", argv[0], argv[0]);
+           "  (e.g. %s -u eng.traineddata tmp/eng.)\n\n", argv[0], argv[0]);
+    printf("Usage for listing the network information\n"
+           "  %s -l traineddata_file\n"
+           "  (e.g. %s -l eng.traineddata)\n\n", argv[0], argv[0]);
     printf(
         "Usage for listing directory of components:\n"
-        "  %s -d traineddata_file\n",
+        "  %s -d traineddata_file\n\n",
         argv[0]);
     printf(
         "Usage for compacting LSTM component to int:\n"