Update README.md

README.md

@@ -20,7 +20,7 @@ This modelization is close to [tblard/tf-allocine](https://huggingface.co/tblard
 Dataset
 -------
 
-The dataset is composed of XXX,XXX reviews for training and X,XXX reviews for the test coming from Amazon, and respectively XXX,XXX and X,XXX critics from [Allocine website](https://www.allocine.fr/). The dataset is labeled into 5 categories:
+The dataset is composed of 204,993 reviews for training and 4,999 reviews for the test coming from Amazon, and respectively 235,516 and 4,729 critics from [Allocine website](https://www.allocine.fr/). The dataset is labeled into 5 categories:
 * 1 star: represents a very bad appreciation,
 * 2 stars: bad appreciation,
 * 3 stars: neutral appreciation,