6 Oct 2025
Days 7 to 9:
I moved on to the NLP part where I learned:
- How the TextVectorization layer works in depth
- Where to place it: in the model or in my pre-processing pipeline - The different types of vectorization (n_gram, one-word and TF-IDF)
(1/3)
1
3
67
6 Oct 2025
Jour 7 à 9 :
Je suis passé à la partie NLP oû j'ai appris:
- Comment fonctionner en profondeur le layer TextVectorization
- Oû le placer : dans le modèle ou bien dans ma pipeline de pre-processing
- Les différents type de vectorization (n_gram, one-word et TF-IDF)
(1/3)
3
1
15
3,518
11 Nov 2022
Going further, you can even specify your own preprocessing layers.
Let's say one of your feature is a text paragraph, and you want to encode it as a TF-IDF vector to be concatenated with your other features. Easy! Create a TextVectorization layer and pass it to the FeatureSpace.
1
3
23
26 Feb 2022
By corpus, do you mean the vocabulary of the tokenizer? Were you using a TextVectorization layer or something else? What did you intend to use the SavedModel for afterwards?
3 Feb 2022
Congrats for releasing TF 2.8 🎉
Nice updates about the TextVectorization layer. No more custom functions for filtering off the string punctuation 👍🏻
2 Feb 2022
TensorFlow 2.8 is out now! 🥳
Congrats to all team members are contributors involved in this release!
github.com/tensorflow/tensor…
1
12
20 Jan 2022
Last call :) You could still join and listen to @DmitryKan & @sophiaalthammer presenting live today at #OpenNLP meetup meetup.com/de-DE/open-nlp-me… #opensource #python #nlproc #machinelearning #textvectorization #densepassageretrieval
9
7
17 Nov 2021
My colleagues from the Keras team Matt Watson and Chen Qian just published a great intro to NLP with Keras and @huggingface. Featuring the new TextVectorization layer in Keras for text preprocessing.
youtu.be/gZIP-_2XYMM
18
119
5 Sep 2021
Natural Language Processing Text Vectorization Approaches: ai.plainenglish.io/natural-l… #MachineLearning #DeepLearning #TextVectorization #MachineIntelligence #ArtificialIntelligence
2
28 Jun 2021
I tried but unfortunately the feature did not make the TensorFlow 2.6 branch cut. Definitely in TF 2.7.
context:
keras.layers.StringLookup and TextVectorization have an output_mode argument for "one_hot", "multi_hot", etc encoding. It is a bit silly that Discretization does not.
2
30 Apr 2021
TensorFlowのTextVectorization Layerめっちゃ便利だけど、split内でmacabとかのtokenizerでsplitできたらさらに便利になりそうだけど、無理なのかな・・・（ちょっといじった感じできなかった）
tensorflow.org/api_docs/pyth…
1
4
6 Feb 2021
Pretrained Word Embeddings using SpaCy and Keras TextVectorization by Josh Johnson buff.ly/39SK6Te
3
20
5 Jan 2021
TextVectorization、なかなか便利やんけ
Tensorflow2.1で追加されたTextVectorizationで文字列から直接訓練できるようになる｜Roy @aflc_jp #note note.com/hroy/n/ncb146928141…
4 Jan 2021
Here's a word-level text generation example with LSTM, starting from raw text files, in less than 50 lines of Keras & TensorFlow. colab.research.google.com/dr…
1
9 Nov 2020
@TensorFlow the new Keras TextVectorization layer is amazing but I'd love to be able to use it in TF Lite and TF JS for our production app. Any chance you could provide a timeline for the support of this layer? Thanks!
2
23 Jul 2020
I'm really liking the new TextVectorization layer. Definitely feels like the future of deployment. Packaging preprocessing logic inside your model simplifies serving & converting py models to other languages like TF.js.
Links: keras.io/examples/nlp/text_c…
tensorflow.org/api_docs/pyth…
1
17
65
9 Jul 2020
Can we integrate @huggingface tokenizers as the preferred tokenization method in TextVectorization layer? This would be really helpful... Last time I tried it in TF 2.1 any custom tokenization function had to deal with tf string type, so I failed miserably 🙂
2
10 Jun 2020
I'm really liking the new TextVectorization layer. Definitely feels like the future of deployment. Packaging preprocessing logic inside your model simplifies serving & converting Python models to other languages like TF.js.
Links: keras.io/examples/nlp/text_c…
tensorflow.org/api_docs/pyth…
1
25
120
18 May 2020
Using pre-trained word embeddings with Keras: buff.ly/3dOQ3zJ
Feat. TextVectorization layer
5
5 May 2020
Using pre-trained word embeddings with Keras: colab.research.google.com/dr…
Feat. TextVectorization layer
6
38
210
28 Apr 2020
Day 045 of #100DaysOfMLCode:
Building a custom search engine from scratch in TF with @fchollet's `TextVectorization` example wouldn't be too difficult. (You'd prefer starting with BERT or a BERT fork, but...)
1. Crawl the web, scraping with BeautifulSoup.
1
2
2