This is an automated email from the ASF dual-hosted git repository.
git-site-role pushed a commit to branch asf-site
in repository https://gitbox.apache.org/repos/asf/groovy-dev-site.git
The following commit(s) were added to refs/heads/asf-site by this push:
new a0c5459 2025/02/27 19:13:12: Generated dev website from
groovy-website@44039c5
a0c5459 is described below
commit a0c54594da24bd7f2c81a600ea42304d7b971d28
Author: jenkins <[email protected]>
AuthorDate: Thu Feb 27 19:13:12 2025 +0000
2025/02/27 19:13:12: Generated dev website from groovy-website@44039c5
---
blog/wayang-tensorflow.html | 25 +++++++++++++++----------
1 file changed, 15 insertions(+), 10 deletions(-)
diff --git a/blog/wayang-tensorflow.html b/blog/wayang-tensorflow.html
index 59abb33..3d39b54 100644
--- a/blog/wayang-tensorflow.html
+++ b/blog/wayang-tensorflow.html
@@ -102,7 +102,10 @@ Java test in the Apache Wayang
</div>
</div>
<div class="paragraph">
-<p>Now we can define a helper method to convert from our test and training CSV
files into our data types.</p>
+<p>Now we can create a helper method to define the operators we’ll use to
+convert from our test and training CSV files into our datasets.
+Operators are the chunks of work that can be allocated to our processing
platform.
+Ultimately, we’ll have a graph of operators that form our plan of
work.</p>
</div>
<div class="listingblock">
<div class="content">
@@ -112,8 +115,8 @@ Java test in the Apache Wayang
new Tuple(it[0..-2]*.toFloat() as float[], LABEL_MAP[it[-1]])
}, String, Tuple)
- var mapData = new MapOperator<>(tuple -> (float[]) tuple.field0,
Tuple, float[]) // <b class="conum">(3)</b>
- var mapLabel = new MapOperator<>(tuple -> (Integer) tuple.field1,
Tuple, Integer) // <b class="conum">(3)</b>
+ var mapData = new MapOperator<>(tuple -> tuple.field0, Tuple,
float[]) // <b class="conum">(3)</b>
+ var mapLabel = new MapOperator<>(tuple -> tuple.field1, Tuple,
Integer) // <b class="conum">(3)</b>
if (random) {
Random r = new Random()
@@ -140,7 +143,7 @@ Java test in the Apache Wayang
<p><code>line2tupleOp</code> converts a line into a Tuple containing our
<code>float[]</code> data in <code>field0</code> and an <code>Integer</code>
label in <code>field1</code></p>
</li>
<li>
-<p>We also have <code>mapData</code> and <code>mapLabe</code> operators for
+<p>We also have <code>mapData</code> and <code>mapLabel</code> operators for
getting the two parts from our Tuple</p>
</li>
<li>
@@ -173,7 +176,7 @@ Operator testLabel = testSource.field1</code></pre>
</div>
</div>
<div class="paragraph">
-<p>Next up we’ll define a model for our deep-learning network.
+<p>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).</p>
</div>
@@ -188,7 +191,9 @@ one or more hidden layers, and outputs (in this case,
labels).</p>
</div>
<div class="paragraph">
<p>We’ll have 4 inputs going to 32 hidden nodes to 3 outputs
-with Sigmoid activation.</p>
+with Sigmoid activation. These classes are all platform-agnostic.
+Nowhere here do we mention TensorFlow or use any TensorFlow
+classes.</p>
</div>
<div class="listingblock">
<div class="content">
@@ -199,8 +204,8 @@ DLModel model = new DLModel(l2)</code></pre>
</div>
</div>
<div class="paragraph">
-<p>We define an operator, providing some needed options, that will do
-our training.</p>
+<p>We define a platform-agnostic deep learning training operator,
+providing some needed options, that will do our training.</p>
</div>
<div class="listingblock">
<div class="content">
@@ -239,7 +244,7 @@ var groundTruthSink = createCollectingSink(groundTruth,
Integer)</code></pre>
</div>
</div>
<div class="paragraph">
-<p>With operators defined, let’s connect them together:</p>
+<p>With operators defined, let’s connect them together (define our graph
of work):</p>
</div>
<div class="listingblock">
<div class="content">
@@ -253,7 +258,7 @@ testLabel.connectTo(0, groundTruthSink, 0)</code></pre>
</div>
</div>
<div class="paragraph">
-<p>Let’s now define a plan and execute it:</p>
+<p>Let’s now place everything in a plan and execute it:</p>
</div>
<div class="listingblock">
<div class="content">
