François Chollet on Twitter: "The new TextVectorization layer in Keras enables you to put your text preprocessing directly inside your models, in a way that is automatically serializable and deployable. You won't have to worry about separately keeping track of your vocabulary again. Here's how to use it:… https://t.co/83DfnQLB77"

Country	Code	For customers of
United States	40404	(any)
Canada	21212	(any)
United Kingdom	86444	Vodafone, Orange, 3, O2
Brazil	40404	Nextel, TIM
Haiti	40404	Digicel, Voila
Ireland	51210	Vodafone, O2
India	53000	Bharti Airtel, Videocon, Reliance
Indonesia	89887	AXIS, 3, Telkomsel, Indosat, XL Axiata
Italy	4880804	Wind
3424486444	Vodafone
» See SMS short codes for other countries

The new TextVectorization layer in Keras enables you to put your text preprocessing directly inside your models, in a way that is automatically serializable and deployable. You won't have to worry about separately keeping track of your vocabulary again. Here's how to use it:pic.twitter.com/62PNcd9Hxb

7:32 PM - 21 Nov 2019

198 Retweets
876 Likes

10 replies 198 retweets 876 likes

1. 에그‏ @eggie5 21 Nov 2019
  Replying to @fchollet
  
  these are native TF graph ops?
  Thanks. Twitter will use this to make your timeline better. Undo
  
  Undo
1. Kartik‏ @ayyar 21 Nov 2019
  Replying to @fchollet
  
  Is there a way to use this with sentence piece ?
  
  4 likes
  Thanks. Twitter will use this to make your timeline better. Undo
  
  Undo
1. zubairahmed_ai‏ @zubairahmed_ai 21 Nov 2019
  Replying to @fchollet
  
  Read yesterday in Word Embedding section of @aureliengeron's new book and liked it, very well explained
  
  2 likes
  Thanks. Twitter will use this to make your timeline better. Undo
  
  Undo
1. Rayees Dar‏ @rayees_dar88 21 Nov 2019
  Replying to @fchollet
  
  That 'text_dataset' is the full Dataset? And as a string?
  
  1 like
  Thanks. Twitter will use this to make your timeline better. Undo
  
  Undo
1. Vaibhav Sharma‏ @vaibshar2014 22 Nov 2019
  Replying to @fchollet
  
  Is there any change to Tokenizer as well?
  Thanks. Twitter will use this to make your timeline better. Undo
  
  Undo
1. xPr06lem TG‏ @Pr06lemChild 22 Nov 2019
  Replying to @fchollet
  
  I’m guessing this is after tokenization?
  Thanks. Twitter will use this to make your timeline better. Undo
  
  Undo
1. Piero Casusol‏ @PieroCV8 22 Nov 2019
  Replying to @fchollet
  
  Amazing, François... One question, yesterday i watched the tf.text presentation on Tensorflow World event. This new layer is related to Ragged Tensors?. And just for curiosity, is possible to apply BERT preprocessing in tf.text to this layer? Thank you
  Thanks. Twitter will use this to make your timeline better. Undo
  
  Undo
1. Show replies
1. Pedro Cardoso‏ @pedroKarudoso 22 Nov 2019
  Replying to @fchollet
  
  How about having this on tf.text?
  Thanks. Twitter will use this to make your timeline better. Undo
  
  Undo