Update README.md

README.md

@@ -7,7 +7,7 @@ Natural Language Processing tasks have been widely studied in the context of big
 - Short texts contains a limited amount of contextual cues;
 - The lack of a unified evaluation framework makes it hard to compare different models.
 
-In this project I am going to use **BERTweet**, a large-scale language model pretrained on English tweets, and I will evaluate its performances on three different NLP tasks:
+In this project I used **BERTweet**, a large-scale language model pretrained on English tweets, and I evaluated its performances on three different NLP tasks:
 
 - *Emotion Recognition*, which consists in recognizing the emotion evoked in a text (a tweet in this case);
 - *Hate Speech Detection*, which consists in understanding whether a tweet is hateful or not towards certain target communities;