This is an automated email from the ASF dual-hosted git repository.
paulk pushed a commit to branch asf-site
in repository https://gitbox.apache.org/repos/asf/groovy-website.git
The following commit(s) were added to refs/heads/asf-site by this push:
new 31d42af draft blog post (cont'd)
31d42af is described below
commit 31d42afa759f4d25abffd1112dec5648dba5e287
Author: Paul King <[email protected]>
AuthorDate: Fri Feb 28 00:08:45 2025 +1000
draft blog post (cont'd)
---
site/src/site/blog/wayang-tensorflow.adoc | 40 +++++++++++++++++++++++++------
1 file changed, 33 insertions(+), 7 deletions(-)
diff --git a/site/src/site/blog/wayang-tensorflow.adoc
b/site/src/site/blog/wayang-tensorflow.adoc
index f41fe65..c58c077 100644
--- a/site/src/site/blog/wayang-tensorflow.adoc
+++ b/site/src/site/blog/wayang-tensorflow.adoc
@@ -84,7 +84,18 @@ Operator testData = testSource.field0
Operator testLabel = testSource.field1
----
-We'll define a model.
+Next up we'll define a model for our deep-learning network.
+Recall that such networks have inputs (the features),
+one or more hidden layers, and outputs (in this case, labels).
+
+image:img/deep_network.png[Iris neural net layers,width=600]
+
+The nodes can be activated by linear or non-linear functions.
+
+image:img/deep_node.png[Neural net node,width=600]
+
+We'll have 4 inputs going to 32 hidden nodes to 3 outputs
+with Sigmoid activation.
[source,groovy]
----
@@ -94,6 +105,9 @@ Op l2 = new Linear(32, 3, true).with(s1)
DLModel model = new DLModel(l2)
----
+We define an operator, providing some needed options, that will do
+our training.
+
[source,groovy]
----
Op criterion = new CrossEntropyLoss(3).with(
@@ -104,27 +118,34 @@ Optimizer optimizer = new Adam(0.1f) // optimizer with
learning rate
int batchSize = 45
int epoch = 10
var option = new DLTrainingOperator.Option(criterion, optimizer, batchSize,
epoch)
-option.setAccuracyCalculation(new Mean(0).with(new
Cast(Op.DType.FLOAT32).with(new Eq().with(
- new ArgMax(1).with(new Input(Input.Type.PREDICTED, Op.DType.FLOAT32)),
- new Input(Input.Type.LABEL, Op.DType.INT32)
+option.setAccuracyCalculation(new Mean(0).with(
+ new Cast(Op.DType.FLOAT32).with(new Eq().with(
+ new ArgMax(1).with(new Input(Input.Type.PREDICTED, Op.DType.FLOAT32)),
+ new Input(Input.Type.LABEL, Op.DType.INT32)
))))
var trainingOp = new DLTrainingOperator<>(model, option, float[], Integer)
----
+Now we'll define a few more operators to work out and collect results:
+
[source,groovy]
----
var predictOp = new PredictOperator<>(float[], float[])
-/* map to label */
-var bestFitOp = new MapOperator<>(array -> array.toList().indexed().max{
it.value }.key, float[], Integer)
+var bestFitOp = new MapOperator<>(array ->
+ array.indexed().max{ it.value }.key, float[], Integer)
-/* sink */
var predicted = []
var predictedSink = createCollectingSink(predicted, Integer)
var groundTruth = []
var groundTruthSink = createCollectingSink(groundTruth, Integer)
+----
+With operators defined, let's connect them together:
+
+[source,groovy]
+----
trainData.connectTo(0, trainingOp, 0)
trainLabel.connectTo(0, trainingOp, 1)
trainingOp.connectTo(0, predictOp, 0)
@@ -132,7 +153,12 @@ testData.connectTo(0, predictOp, 1)
predictOp.connectTo(0, bestFitOp, 0)
bestFitOp.connectTo(0, predictedSink, 0)
testLabel.connectTo(0, groundTruthSink, 0)
+----
+Let's now define a plan and execute it:
+
+[source,groovy]
+----
var wayangPlan = new WayangPlan(predictedSink, groundTruthSink)
new WayangContext().with {
