Add basic definition of models, Needs to be trained and tested

Trained a lenet1 model Add All Weights Add unzip utility function, fix linux build Completed everything Fix build, Style fixes next Use force local for windows while unzipping tar files Use force local for windows while unzipping tar files Added Utility Function Added Utility Function Style Fix Fix Typo causing build error in windows Fix Typo causing build error in windows Fix const issue for windows Extract in data folder Reposition force local This should work Print Path in windows for debugging Print Path in windows for debugging Print Path in windows for debugging Stip components of tar Stip components of tar strip componenets Initial Definition Add List Dir utility function Add List Dir utility function Squash this, annotation read Yay, we can read xml now
mlpack · May 29, 2020 · 3353e2e · 3353e2e
1 parent e70d009
commit 3353e2e
Show file tree

Hide file tree

Showing 12 changed files with 445 additions and 30 deletions.
diff --git a/.ci/linux-steps.yaml b/.ci/linux-steps.yaml
@@ -45,7 +45,7 @@ steps:
   displayName: 'Build models'
 
 # Run CTests.
-- script: cd build/tests/ && sudo CTEST_OUTPUT_ON_FAILURE=1 ctest -R UtilsTest
+- script: cd build/tests/ && sudo CTEST_OUTPUT_ON_FAILURE=1 ctest -T Test .
   displayName: 'Run tests via ctest'
 
 # Publish test results to Azure Pipelines

diff --git a/.ci/macos-steps.yaml b/.ci/macos-steps.yaml
@@ -37,7 +37,7 @@ steps:
   displayName: 'Build models'
 
 # Run CTests.
-- script: cd build/tests/ && sudo CTEST_OUTPUT_ON_FAILURE=1 ctest -R UtilsTest
+- script: cd build/tests/ && sudo CTEST_OUTPUT_ON_FAILURE=1 ctest -T Test .
   displayName: 'Run tests via ctest'
 
 # Publish test results to Azure Pipelines

diff --git a/.ci/windows-steps.yaml b/.ci/windows-steps.yaml
@@ -134,7 +134,7 @@ steps:
 # Run tests via ctest.
 - bash: |
     cd build/tests
-    CTEST_OUTPUT_ON_FAILURE=1 ctest -T Test -C Release -R UtilsTest
+    CTEST_OUTPUT_ON_FAILURE=1 ctest -T Test -C Release .
   displayName: 'Run tests via ctest'
 
 # Publish test results to Azure Pipelines

diff --git a/.gitignore b/.gitignore
@@ -4,5 +4,8 @@ xcode*
 .idea
 cmake-build-*
 *.csv
+*.tar
+*.zip
+*.tar.gz
 .travis/configs.hpp
 Testing/*
diff --git a/data/annotations/2007_000027.xml b/data/annotations/2007_000027.xml
@@ -0,0 +1,63 @@
+<annotation>
+	<folder>VOC2012</folder>
+	<filename>2007_000027.jpg</filename>
+	<source>
+		<database>The VOC2007 Database</database>
+		<annotation>PASCAL VOC2007</annotation>
+		<image>flickr</image>
+	</source>
+	<size>
+		<width>486</width>
+		<height>500</height>
+		<depth>3</depth>
+	</size>
+	<segmented>0</segmented>
+	<object>
+		<name>person</name>
+		<pose>Unspecified</pose>
+		<truncated>0</truncated>
+		<difficult>0</difficult>
+		<bndbox>
+			<xmin>174</xmin>
+			<ymin>101</ymin>
+			<xmax>349</xmax>
+			<ymax>351</ymax>
+		</bndbox>
+		<part>
+			<name>head</name>
+			<bndbox>
+				<xmin>169</xmin>
+				<ymin>104</ymin>
+				<xmax>209</xmax>
+				<ymax>146</ymax>
+			</bndbox>
+		</part>
+		<part>
+			<name>hand</name>
+			<bndbox>
+				<xmin>278</xmin>
+				<ymin>210</ymin>
+				<xmax>297</xmax>
+				<ymax>233</ymax>
+			</bndbox>
+		</part>
+		<part>
+			<name>foot</name>
+			<bndbox>
+				<xmin>273</xmin>
+				<ymin>333</ymin>
+				<xmax>297</xmax>
+				<ymax>354</ymax>
+			</bndbox>
+		</part>
+		<part>
+			<name>foot</name>
+			<bndbox>
+				<xmin>319</xmin>
+				<ymin>307</ymin>
+				<xmax>340</xmax>
+				<ymax>326</ymax>
+			</bndbox>
+		</part>
+	</object>
+</annotation>
diff --git a/data/annotations/2007_000032.xml b/data/annotations/2007_000032.xml
@@ -0,0 +1,63 @@
+<annotation>
+	<folder>VOC2012</folder>
+	<filename>2007_000032.jpg</filename>
+	<source>
+		<database>The VOC2007 Database</database>
+		<annotation>PASCAL VOC2007</annotation>
+		<image>flickr</image>
+	</source>
+	<size>
+		<width>500</width>
+		<height>281</height>
+		<depth>3</depth>
+	</size>
+	<segmented>1</segmented>
+	<object>
+		<name>aeroplane</name>
+		<pose>Frontal</pose>
+		<truncated>0</truncated>
+		<difficult>0</difficult>
+		<bndbox>
+			<xmin>104</xmin>
+			<ymin>78</ymin>
+			<xmax>375</xmax>
+			<ymax>183</ymax>
+		</bndbox>
+	</object>
+	<object>
+		<name>aeroplane</name>
+		<pose>Left</pose>
+		<truncated>0</truncated>
+		<difficult>0</difficult>
+		<bndbox>
+			<xmin>133</xmin>
+			<ymin>88</ymin>
+			<xmax>197</xmax>
+			<ymax>123</ymax>
+		</bndbox>
+	</object>
+	<object>
+		<name>person</name>
+		<pose>Rear</pose>
+		<truncated>0</truncated>
+		<difficult>0</difficult>
+		<bndbox>
+			<xmin>195</xmin>
+			<ymin>180</ymin>
+			<xmax>213</xmax>
+			<ymax>229</ymax>
+		</bndbox>
+	</object>
+	<object>
+		<name>person</name>
+		<pose>Rear</pose>
+		<truncated>0</truncated>
+		<difficult>0</difficult>
+		<bndbox>
+			<xmin>26</xmin>
+			<ymin>189</ymin>
+			<xmax>44</xmax>
+			<ymax>238</ymax>
+		</bndbox>
+	</object>
+</annotation>
diff --git a/dataloader/dataloader.hpp b/dataloader/dataloader.hpp
@@ -10,10 +10,13 @@
  * http://www.opensource.org/licenses/BSD-3-Clause for more information.
  */
 #include <mlpack/core/data/scaler_methods/min_max_scaler.hpp>
+#include <boost/property_tree/xml_parser.hpp>
 #include <mlpack/core/math/shuffle_data.hpp>
 #include <mlpack/core/data/split_data.hpp>
-#include <mlpack/prereqs.hpp>
+#include <boost/property_tree/ptree.hpp>
 #include <dataloader/datasets.hpp>
+#include <mlpack/prereqs.hpp>
+#include <boost/foreach.hpp>
 #include <mlpack/core.hpp>
 #include <utils/utils.hpp>
 
@@ -117,6 +120,28 @@ class DataLoader
                    std::vector<std::string>(),
                const double augmentationProbability = 0.2);
 
+  /**
+   * Loads object detection dataset. It requires a single annotation file in xml format.
+   * Each XML file should correspond to a single image in images folder.
+   *
+   * XML file should containg the following :
+   * 1. Each XML file should be wrapped in annotation tag.
+   * 2. Filename of image in images folder will be depicted by filename tag.
+   * 3. Object tag depicting characteristics of bounding box.
+   * 4. Each object tag should contain name tag i.e. class of the object.
+   * 5. Each object tag should contain bndbox tag containing xmin, ymin, xmax, ymax.
+   *
+   * NOTE : Labels are assigned using lexicographically. Set verbose to 1 to print labels
+   * and their corresponding class.
+   *
+   * @param pathToAnnotations Path to the folder containg xml type annotation files.
+   * @param pathToImages Path to folder containing images corresponding to annotations.
+   * @param absolutePath Boolean to determine if absolute path is used. Defaults to false.
+   */
+  void LoadObjectDetectionDataset(const std::string& pathToAnnotations,
+                                  const std::string& pathToImages,
+                                  const bool absolutePath = false);
+
   //! Get the training dataset features.
   DatasetX TrainFeatures() const { return trainFeatures; }
 
@@ -179,11 +204,30 @@ class DataLoader
    */
   void DownloadDataset(const std::string& dataset)
   {
+    if (datasetMap[dataset].zipFile && (!Utils::PathExists(
+        datasetMap[dataset].trainPath) ||
+        !Utils::PathExists(datasetMap[dataset].testPath)))
+    {
+      Utils::DownloadFile(datasetMap[dataset].datasetURL,
+          datasetMap[dataset].datasetPath, dataset + "_training_data.",
+          false, false, datasetMap[dataset].serverName,
+          datasetMap[dataset].zipFile);
+
+      if (!Utils::CompareCRC32(datasetMap[dataset].datasetPath,
+          datasetMap[dataset].datasetHash))
+      {
+        mlpack::Log::Fatal << "Corrupted Data for " << dataset <<
+            " downloaded." << std::endl;
+      }
+
+      return;
+    }
+
     if (!Utils::PathExists(datasetMap[dataset].trainPath))
     {
-      Utils::DownloadFile(datasetMap[dataset].trainDownloadUrl,
+      Utils::DownloadFile(datasetMap[dataset].trainDownloadURL,
           datasetMap[dataset].trainPath, dataset + "_training_data.",
-          false);
+          false, false, datasetMap[dataset].serverName);
 
       if (!Utils::CompareCRC32(datasetMap[dataset].trainPath,
           datasetMap[dataset].trainHash))
@@ -192,11 +236,12 @@ class DataLoader
             dataset << " downloaded." << std::endl;
       }
     }
+
     if (!Utils::PathExists(datasetMap[dataset].testPath))
     {
-      Utils::DownloadFile(datasetMap[dataset].trainDownloadUrl,
+      Utils::DownloadFile(datasetMap[dataset].trainDownloadURL,
           datasetMap[dataset].testPath, dataset + "_testing_data.",
-          false);
+          false, false, datasetMap[dataset].serverName);
 
       if (!Utils::CompareCRC32(datasetMap[dataset].testPath,
           datasetMap[dataset].testHash))

diff --git a/dataloader/dataloader_impl.hpp b/dataloader/dataloader_impl.hpp
@@ -106,13 +106,6 @@ template<
     arma::mat trainDataset, validDataset;
     data::Split(dataset, trainDataset, validDataset, ratio, shuffle);
 
-    if (useScaler)
-    {
-      scaler.Fit(trainDataset);
-      scaler.Transform(trainDataset, trainDataset);
-      scaler.Transform(validDataset, validDataset);
-    }
-
     trainFeatures = trainDataset.rows(WrapIndex(startInputFeatures,
         trainDataset.n_rows), WrapIndex(endInputFeatures,
         trainDataset.n_rows));
@@ -125,10 +118,16 @@ template<
         validDataset.n_rows), WrapIndex(endInputFeatures,
         validDataset.n_rows));
 
-    validLabels = trainDataset.rows(WrapIndex(startPredictionFeatures,
+    validLabels = validDataset.rows(WrapIndex(startPredictionFeatures,
         validDataset.n_rows), WrapIndex(endPredictionFeatures,
         validDataset.n_rows));
 
+    if (useScaler)
+    {
+      scaler.Fit(trainFeatures);
+      scaler.Transform(trainFeatures, trainFeatures);
+      scaler.Transform(validFeatures, validFeatures);
+    }
     // TODO : Add support for augmentation here.
     mlpack::Log::Info << "Training Dataset Loaded." << std::endl;
   }
@@ -145,4 +144,37 @@ template<
   }
 }
 
+template<
+  typename DatasetX,
+  typename DatasetY,
+  class ScalerType
+> void DataLoader<
+    DatasetX, DatasetY, ScalerType
+>::LoadObjectDetectionDataset(const std::string& pathToAnnotations,
+                              const std::string& pathToImages,
+                              const bool absolutePath)
+{
+  std::vector<boost::filesystem::path> annotationsDirectory, imagesDirectory;
+
+  // Fill the directory.
+  Utils::ListDir(pathToAnnotations, annotationsDirectory, absolutePath);
+  Utils::ListDir(pathToImages, imagesDirectory, absolutePath);
+
+  // Read the xml file.
+  for (boost::filesystem::path annotationFile : annotationsDirectory)
+  {
+    // Read the xml file.
+    boost::property_tree::ptree annotation;
+    std::cout << annotationFile.string() << std::endl;
+    boost::property_tree::read_xml(annotationFile.string(), annotation);
+
+    // Read properties inside annotation file.
+    BOOST_FOREACH (boost::property_tree::ptree::value_type const& object,
+        annotation.get_child("annotation.object"))
+    {
+      std::cout << object.first << std::endl;
+    }
+  }
+}
+
 #endif