Update nmt_with_attention

For TF2.5 - Use the TextVectorization layer. - Use the AdditiveAttention layer. - tf.function the translate loop for text->text export. - Add more inline explanations, and sanity checks. - Add shape assertions throughout the code to make it easier to follow. Fixes: tensorflow/tensorflow#38248 Fixes: tensorflow/tensorflow#39654 See also: tensorflow/tensorflow#49237 PiperOrigin-RevId: 370250185
tensorflow · May 25, 2021 · 0d2a2fa · 0d2a2fa
1 parent a20cc17
commit 0d2a2fa
Show file tree

Hide file tree

Showing 6 changed files with 2,430 additions and 458 deletions.
diff --git a/docs/tutorials/BUILD b/docs/tutorials/BUILD
@@ -29,7 +29,7 @@ tf_org_notebook_test(
     deps = [
         "//third_party/py/matplotlib",
         #  numpy dep,
-        "//third_party/py/sklearn",
+        "//third_party/py/tensorflow_text",
     ],
 )
 

diff --git a/docs/tutorials/images/attention_equation_1.jpg b/docs/tutorials/images/attention_equation_1.jpg
diff --git a/docs/tutorials/images/attention_equation_2.jpg b/docs/tutorials/images/attention_equation_2.jpg
diff --git a/docs/tutorials/images/attention_equation_3.jpg b/docs/tutorials/images/attention_equation_3.jpg
diff --git a/docs/tutorials/images/attention_equation_4.jpg b/docs/tutorials/images/attention_equation_4.jpg