Changeset 17489 for branches/3040_VectorBasedGP/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/TensorFlowConstantOptimizationEvaluator.cs

Timestamp:

04/01/20 15:49:03 (4 years ago)

Author:

pfleck

Message:

#3040 Added version with explicit array shapes for explicit broadcasting.

File:

• branches/3040_VectorBasedGP/HeuristicLab.Problems.DataAnalysis.Symbolic.Regression/3.4/SingleObjective/Evaluators/TensorFlowConstantOptimizationEvaluator.cs (modified) (4 diffs)

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

@@ -19 +19 @@
  */
 #endregion
+
+#define EXPLICIT_SHAPE
 
 using System;
@@ -94 +96 @@
       CancellationToken cancellationToken = default(CancellationToken), EvaluationsCounter counter = null) {
 
-      var vectorVariables = tree.IterateNodesBreadth()
-        .OfType<VariableTreeNodeBase>()
-        .Where(node => problemData.Dataset.VariableHasType<DoubleVector>(node.VariableName))
-        .Select(node => node.VariableName);
-
-      int? vectorLength = null;
-      if (vectorVariables.Any()) {
-        vectorLength = vectorVariables.Select(var => problemData.Dataset.GetDoubleVectorValues(var, rows)).First().First().Count;
-      }
       int numRows = rows.Count();
+      var variableLengths = problemData.AllowedInputVariables.ToDictionary(
+        var => var,
+        var => {
+          if (problemData.Dataset.VariableHasType<double>(var)) return 1;
+          if (problemData.Dataset.VariableHasType<DoubleVector>(var)) return problemData.Dataset.GetDoubleVectorValue(var, 0).Count;
+          throw new NotSupportedException($"Type of variable {var} is not supported.");
+        });
 
       bool success = TreeToTensorConverter.TryConvert(tree,
-        numRows, vectorLength,
+        numRows, variableLengths,
         updateVariableWeights, applyLinearScaling,
         out Tensor prediction,
         out Dictionary<Tensor, string> parameters, out List<Tensor> variables/*, out double[] initialConstants*/);
 
-      var target = tf.placeholder(tf.float64, name: problemData.TargetVariable);
-      int samples = rows.Count();
+#if EXPLICIT_SHAPE
+      var target = tf.placeholder(tf.float64, new TensorShape(numRows, 1), name: problemData.TargetVariable);
+#endif
       // mse
-      var costs = tf.reduce_sum(tf.square(prediction - target)) / (2.0 * samples);
+      var costs = tf.reduce_sum(tf.square(target - prediction)) / (2.0 * numRows);
       var optimizer = tf.train.GradientDescentOptimizer((float)learningRate).minimize(costs);
 
@@ -124 +125 @@
         if (problemData.Dataset.VariableHasType<double>(variableName)) {
           var data = problemData.Dataset.GetDoubleValues(variableName, rows).ToArray();
-          if (vectorLength.HasValue) {
-            var vectorData = new double[numRows][];
-            for (int i = 0; i < numRows; i++)
-              vectorData[i] = Enumerable.Repeat(data[i], vectorLength.Value).ToArray();
-            variablesFeed.Add(variable, np.array(vectorData));
-          } else
-            variablesFeed.Add(variable, np.array(data, copy: false));
+          //if (vectorLength.HasValue) {
+          //  var vectorData = new double[numRows][];
+          //  for (int i = 0; i < numRows; i++)
+          //    vectorData[i] = Enumerable.Repeat(data[i], vectorLength.Value).ToArray();
+          //  variablesFeed.Add(variable, np.array(vectorData));
+          //} else
+          variablesFeed.Add(variable, np.array(data, copy: false).reshape(numRows, 1));
           //} else if (problemData.Dataset.VariableHasType<string>(variableName)) {
           //  variablesFeed.Add(variable, problemData.Dataset.GetStringValues(variableName, rows));
@@ -140 +141 @@
       }
       var targetData = problemData.Dataset.GetDoubleValues(problemData.TargetVariable, rows).ToArray();
-      variablesFeed.Add(target, np.array(targetData, copy: false));
-
+      variablesFeed.Add(target, np.array(targetData, copy: false).reshape(numRows, 1));
 
       using (var session = tf.Session()) {
         session.run(tf.global_variables_initializer());
+
+        // https://github.com/SciSharp/TensorFlow.NET/wiki/Debugging
+        tf.train.export_meta_graph(@"C:\temp\TFboard\graph.meta", as_text: false);
 
         Trace.WriteLine("Weights:");