Merge branch 'gpu' of git@git.elphel.com:Elphel/imagej-elphel.git into gpu

src/main/java/TensorflowExamplePlugin.java

@@ -12,6 +12,7 @@ import org.tensorflow.SavedModelBundle;
 import org.tensorflow.OperationBuilder;
 import org.tensorflow.Shape;
 import org.tensorflow.Output;
+import org.tensorflow.Operation;
 
 import java.util.ArrayList;
 import java.util.Collection;
@@ -51,7 +52,9 @@ public class TensorflowExamplePlugin
 {
 
     public final static String EXPORTDIR = "/home/oleg/GIT/python3-imagej-tiff/data_sets/tf_data_5x5_main_13_heur/exportdir";
-    public final static String PB_TAG = "model_pb";
+    // tf.saved_model.tag_constants.SERVING = "serve"
+    public final static String SERVING = "serve";
+    
 
     public static void run()
     {
@@ -115,57 +118,60 @@ public class TensorflowExamplePlugin
 
         final Graph smpb;
 
+        // init for variable?
         float [][] rv_stage1_out = new float[78408][32];
         
-        // from: infer_qcds_01.py
+        // from infer_qcds_01.py
         float [][] img_corr2d = new float[78408][324];
         float [][] img_target = new float[78408][  1];
         int     [] img_ntile  = new   int[78408];
         
-        // init ntile
+        // init ntile for testing?
         for(int i=0;i<img_ntile.length;i++){
         	img_ntile[i] = i;
         }
-
-        /*
-         * for feed:
-         *     "ph_corr2d":           img_corr2d
-         *     "ph_target_disparity": img_target
-         *     "ph_ntile":            img_ntile
-         *     
-         *     so it will look like:
-         *     
-         *     https://divis.io/2018/01/enterprise-tensorflow-code-examples/ -> 
-         *         https://github.com/DIVSIO/tensorflow_java_cli_example/blob/master/src/main/java/divisio/example/tensorflow/cli/RunRegression.java
-         *     
-         *     sess.runner()
-         *         .feed("ph_corr2d",img_corr2d)
-         *         .feed("ph_target_disparity",img_target)
-         *         .feed("ph_ntile",img_ntile)
-         *         .fetch("Disparity_net/stage1done:0")
-         *         .run()
-         *         .get(0)
-         */
         
-        final SavedModelBundle bundle = SavedModelBundle.load(EXPORTDIR,PB_TAG); 
+        final SavedModelBundle bundle = SavedModelBundle.load(EXPORTDIR,SERVING); 
         
         final List<Tensor<?>> tensorsToClose = new ArrayList<Tensor<?>>(5);
         
         System.out.println("OK");
         
         try {
-
-        	// init variable via constant
-        	Tensor<Float> t = toTensor2DFloat(rv_stage1_out, tensorsToClose);
-        	Output builder_init = bundle.graph().opBuilder("Const", "rv_stage1_out_init").setAttr("dtype", t.dataType()).setAttr("value", t).build().output(0);
+        	
+        	System.out.println("S0:");
+        	// read Variable info test
+        	Operation opr = bundle.graph().operation("rv_stage1_out");
+        	System.out.println(opr.toString());
+        	
+        	System.out.println("S1:");
+        	
+        	// init variable via constant?
+        	Tensor<Float> tsr = toTensor2DFloat(rv_stage1_out, tensorsToClose);
+        	Output builder_init = bundle.graph()
+        			                    .opBuilder("Const", "rv_stage1_out_init")
+        			                    .setAttr  ("dtype", tsr.dataType())
+        			                    .setAttr  ("value", tsr)
+        			                    .build()
+        			                    .output(0);
+        	
+        	System.out.println(builder_init);
         	
         	// variable
-        	OperationBuilder builder2 = bundle.graph().opBuilder("Variable", "rv_stage1_out");
-        	builder2.addInput(builder_init);
+        	OperationBuilder builder2 = bundle.graph().opBuilder("Variable", "rv_stage1_out_extra_variable");
+        											  //.addInput(builder_init);
         	
-        	//Tensor<Float> t = toTensor2DFloat(rv_stage1_out, tensorsToClose);
-        	//builder.setAttr("dtype", t.dataType()).setAttr("shape",t.shape()).build().output(0);
+        	//builder2.
+        	//bundle.graph().opBuilder("Assign", "Assign/" + builder2.op().name()).addInput(variable).addInput(value).build().output(0);
         	
+        	//Tensor<Float> tensorVal = tsr;
+        	//Output oValue = bundle.graph().opBuilder("Const", "rv_stage1_out_2").setAttr("dtype", tensorVal.dataType()).setAttr("value", tensorVal).build().output(0);
+        	//System.out.println(oValue);
+        	//Output oValue = bundle.graph().opBuilder("Variable", "rv_stage1_out").setAttr("value", tensorVal).build().output(0);
+        	//bundle.graph().opBuilder("Assign", "Assign/rv_stage1_out").setAttr("value", tsr).build();
+        	        	
+        	System.out.println("DONE");
+        	        	
         	// stage 1        	
         	bundle.session().runner()
 	                  .feed("ph_corr2d",toTensor2DFloat(img_corr2d, tensorsToClose))
@@ -186,11 +192,12 @@ public class TensorflowExamplePlugin
         	
         	float [] resultValues = (float[]) result.copyTo(new float[78408]);
         	
+        	
         	System.out.println("DONE");
         	
-        } catch (final IllegalStateException ise) {
-        	System.out.println("Very Bad Error (VBE): "+ise);
-        	closeTensors(tensorsToClose);
+        //} catch (final IllegalStateException ise) {
+        //	System.out.println("Very Bad Error (VBE): "+ise);
+        //	closeTensors(tensorsToClose);
         } catch (final NumberFormatException nfe) {
 			//just skip unparsable lines ?!
 		} finally {