connect Kernel and Linear machine to a signal handler (#4287)

* connect Kernel and Linear machine to a signal handler
* using base class methods

src/shogun/machine/KernelMachine.cpp

@@ -6,14 +6,14 @@
  *          Fernando Iglesias, Thoralf Klein
  */
 
+#include <rxcpp/rx-lite.hpp>
 #include <shogun/base/progress.h>
 #include <shogun/io/SGIO.h>
-#include <shogun/labels/RegressionLabels.h>
-#include <shogun/machine/KernelMachine.h>
-
-#include <shogun/kernel/Kernel.h>
 #include <shogun/kernel/CustomKernel.h>
+#include <shogun/kernel/Kernel.h>
 #include <shogun/labels/Labels.h>
+#include <shogun/labels/RegressionLabels.h>
+#include <shogun/machine/KernelMachine.h>
 
 #ifdef HAVE_OPENMP
 #include <omp.h>
@@ -421,10 +421,6 @@ void CKernelMachine::store_model_features()
 
 bool CKernelMachine::train_locked(SGVector<index_t> indices)
 {
-	SG_DEBUG("entering %s::train_locked()\n", get_name())
-	if (!is_data_locked())
-		SG_ERROR("CKernelMachine::train_locked() call data_lock() before!\n")
-
 	/* this is asusmed here */
 	ASSERT(m_custom_kernel==kernel)
 
@@ -443,15 +439,13 @@ bool CKernelMachine::train_locked(SGVector<index_t> indices)
 
 	/* dont do train because model should not be stored (no acutal features)
 	 * and train does data_unlock */
-	bool result=train_machine();
-
+	bool result = CMachine::train_locked();
 	/* remove last col subset of custom kernel */
 	m_custom_kernel->remove_col_subset();
 
 	/* remove label subset after training */
 	m_labels->remove_subset();
 
-	SG_DEBUG("leaving %s::train_locked()\n", get_name())
 	return result;
 }
 

src/shogun/machine/KernelMachine.h

@@ -227,11 +227,12 @@ class CKernelMachine : public CMachine
 		virtual float64_t apply_one(int32_t num);
 
 #ifndef SWIG // SWIG should skip this part
-		/** Trains a locked machine on a set of indices. Error if machine is
-		 * not locked
+
+		/** This precomputes the kernel matrix and stores it
 		 *
-		 * @param indices index vector (of locked features) that is used for training
-		 * @return whether training was successful
+		 *  @param indices index vector (of locked features) that is used for
++		 *  training
+		 *  @return whether training was successful
 		 */
 		virtual bool train_locked(SGVector<index_t> indices);
 

src/shogun/machine/LinearMachine.cpp

@@ -6,10 +6,11 @@
  *          Fernando Iglesias
  */
 
-#include <shogun/machine/LinearMachine.h>
-#include <shogun/labels/RegressionLabels.h>
+#include <rxcpp/rx-lite.hpp>
 #include <shogun/features/DotFeatures.h>
 #include <shogun/labels/Labels.h>
+#include <shogun/labels/RegressionLabels.h>
+#include <shogun/machine/LinearMachine.h>
 #include <shogun/mathematics/eigen3.h>
 
 using namespace shogun;
@@ -166,25 +167,9 @@ void CLinearMachine::compute_bias(CFeatures* data)
 
 bool CLinearMachine::train(CFeatures* data)
 {
-    /* not allowed to train on locked data */
-    if (m_data_locked)
-    {
-    	SG_ERROR("train data_lock() was called, only train_locked() is"
-    		" possible. Call data_unlock if you want to call train()\n",
-    		get_name());
-    }
-
-    if (train_require_labels())
-    {
-    	REQUIRE(m_labels,"No labels given",this->get_name());
-
-    	m_labels->ensure_valid(get_name());
-    }
-
-    bool result = train_machine(data);
-
-    if(m_compute_bias)
-    	compute_bias(data);
+	bool result = CMachine::train(data);
+	if (m_compute_bias)
+		compute_bias(data);
 
-    return result;
+	return result;
 }

src/shogun/machine/Machine.cpp

@@ -40,12 +40,11 @@ CMachine::~CMachine()
 bool CMachine::train(CFeatures* data)
 {
 	/* not allowed to train on locked data */
-	if (m_data_locked)
-	{
-		SG_ERROR("%s::train data_lock() was called, only train_locked() is"
-				" possible. Call data_unlock if you want to call train()\n",
-				get_name());
-	}
+	REQUIRE(
+	    !m_data_locked, "(%s)::train data_lock() was called, only "
+	                    "train_locked() is possible. Call data_unlock if you "
+	                    "want to call train()\n",
+	    get_name());
 
 	if (train_require_labels())
 	{
@@ -66,6 +65,20 @@ bool CMachine::train(CFeatures* data)
 	return result;
 }
 
+bool CMachine::train_locked()
+{
+	/*train machine without any actual features(data is locked)*/
+	REQUIRE(
+	    is_data_locked(),
+	    "Data needs to be locked for training, call data_lock()\n")
+
+	auto sub = connect_to_signal_handler();
+	bool result = train_machine();
+	sub.unsubscribe();
+	reset_computation_variables();
+	return result;
+}
+
 void CMachine::set_labels(CLabels* lab)
 {
     if (lab != NULL)

src/shogun/machine/Machine.h

@@ -159,6 +159,14 @@ class CMachine : public CStoppableSGObject
 		 */
 		virtual bool train(CFeatures* data=NULL);
 
+#ifndef SWIG // SWIG should skip this part
+		/** Trains a locked machine on a set of indices. Error if machine is
+		 * not locked
+		 * @return whether training was successful
+		 */
+		virtual bool train_locked();
+#endif
+
 		/** apply machine to data
 		 * if data is not specified apply to the current features
 		 *