From 2cc019092cde25b9c9bb6d6b770bf57d8ad7eaea Mon Sep 17 00:00:00 2001
From: Adel Benlagra <benlagra.adel@courrier.uqam.ca>
Date: Mon, 13 Nov 2023 17:16:22 -0500
Subject: [PATCH] converted multi_label_classification exemple

---
 .../tensorflow/nlp/multi_label_classification.py  | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

diff --git a/examples/keras_io/tensorflow/nlp/multi_label_classification.py b/examples/keras_io/tensorflow/nlp/multi_label_classification.py
index d6f43314dce..c4ac6014639 100644
--- a/examples/keras_io/tensorflow/nlp/multi_label_classification.py
+++ b/examples/keras_io/tensorflow/nlp/multi_label_classification.py
@@ -2,7 +2,7 @@
 Title: Large-scale multi-label text classification
 Author: [Sayak Paul](https://twitter.com/RisingSayak), [Soumik Rakshit](https://github.com/soumik12345)
 Date created: 2020/09/25
-Last modified: 2020/12/23
+Last modified: 2023/11/13 (conversion to Keras 3 by @ben-ad)
 Description: Implementing a large-scale multi-label text classification model.
 Accelerator: GPU
 """
@@ -29,8 +29,8 @@
 ## Imports
 """
 
-from tensorflow.keras import layers
-from tensorflow import keras
+from keras import layers
+import keras
 import tensorflow as tf
 
 from sklearn.model_selection import train_test_split
@@ -65,7 +65,7 @@
 duplication. Here we notice that our initial dataset has got about 13k duplicate entries.
 """
 
-total_duplicate_titles = sum(arxiv_data["titles"].duplicated())
+total_duplicate_titles = arxiv_data["titles"].duplicated().sum()
 print(f"There are {total_duplicate_titles} duplicate titles.")
 
 """
@@ -144,7 +144,7 @@
 """
 
 terms = tf.ragged.constant(train_df["terms"].values)
-lookup = tf.keras.layers.StringLookup(output_mode="multi_hot")
+lookup = layers.StringLookup(output_mode="multi_hot")
 lookup.adapt(terms)
 vocab = lookup.get_vocabulary()
 
@@ -371,7 +371,10 @@ def plot_result(item):
 
 An important feature of the
 [preprocessing layers provided by Keras](https://keras.io/guides/preprocessing_layers/)
-is that they can be included inside a `tf.keras.Model`. We will export an inference model
+is that they can be included inside a `keras.Model`. Note however that the [TextVectorization 
+layer](https://keras.io/keras_core/api/layers/preprocessing_layers/text/text_vectorization/)
+ uses TensorFlow internally. It cannot be used as part of a compiled computation graph 
+of a model with any backend other than TensorFlow. We will export an inference model
 by including the `text_vectorization` layer on top of `shallow_mlp_model`. This will
 allow our inference model to directly operate on raw strings.