Changeset 17721 for branches/3040_VectorBasedGP/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression

Timestamp:

08/11/20 13:39:48 (4 years ago)

Author:

pfleck

Message:

#3040 First draft of different-vector-length strategies (cut, fill, resample, cycle, ...)

File:

• branches/3040_VectorBasedGP/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/TensorFlowConstantOptimizationEvaluator.cs (modified) (6 diffs)

branches/3040_VectorBasedGP/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/TensorFlowConstantOptimizationEvaluator.cs

@@ -56 +56 @@
     private static readonly TF_DataType DataType = tf.float32;
 
-#region Parameter Properties
+    #region Parameter Properties
     public IFixedValueParameter<IntValue> ConstantOptimizationIterationsParameter {
       get { return (IFixedValueParameter<IntValue>)Parameters[MaximumIterationsName]; }
@@ -63 +63 @@
       get { return (IFixedValueParameter<DoubleValue>)Parameters[LearningRateName]; }
     }
-#endregion
-
-#region Properties
+    #endregion
+
+    #region Properties
     public int ConstantOptimizationIterations {
       get { return ConstantOptimizationIterationsParameter.Value.Value; }
@@ -72 +72 @@
       get { return LearningRateParameter.Value.Value; }
     }
-#endregion
+    #endregion
 
     public TensorFlowConstantOptimizationEvaluator()
@@ -125 +125 @@
       var target = tf.placeholder(DataType, new TensorShape(numRows), name: problemData.TargetVariable);
       // MSE
-      var cost = tf.reduce_mean(tf.square(prediction - target));
+      var cost = tf.reduce_mean(tf.square(target - prediction));
 
       var optimizer = tf.train.AdamOptimizer((float)learningRate);
@@ -144 +144 @@
         if (problemData.Dataset.VariableHasType<double>(variableName)) {
           var data = problemData.Dataset.GetDoubleValues(variableName, rows).Select(x => (float)x).ToArray();
-          variablesFeed.Add(variable, np.array(data, copy: false).reshape(numRows, 1));
+          variablesFeed.Add(variable, np.array(data).reshape(numRows, 1));
         } else if (problemData.Dataset.VariableHasType<DoubleVector>(variableName)) {
           var data = problemData.Dataset.GetDoubleVectorValues(variableName, rows).Select(x => x.Select(y => (float)y).ToArray()).ToArray();
@@ -152 +152 @@
       }
       var targetData = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, rows).Select(x => (float)x).ToArray();
-      variablesFeed.Add(target, np.array(targetData, copy: false));
+      variablesFeed.Add(target, np.array(targetData));