Update README.md

README.md

@@ -3,6 +3,9 @@ library_name: keras
 tags:
 - translation
 license: apache-2.0
+language:
+- en
+- pt
 ---
 # GRU-eng-por
 
@@ -154,19 +157,49 @@ Portuguese translation:
 [start] não faça isso [end]
 --------------------------------------------------
 ```
+## Intended Use
 
+This model was created for research purposes only. Specifically, it was designed to translate sentences from English to Portuguese. 
+We do not recommend any application of this model outside this scope.
 
-# Cite as 🤗
-```
+## Performance Metrics
+
+Accuracy is a crude way to monitor validation-set performance during this task. 
+On average, this model correctly predicts words in the Portuguese sentence: 65%.
+However, next-token accuracy isn't an excellent metric for machine translation models. 
+During inference, you're generating the target sentence from scratch and can't rely on previously generated tokens (a.k.a. 100% correctness does not mean you have a good translator). 
+We would likely use "_BLEU scores_" in real-world machine translation applications to evaluate our models.
+
+## Training Data
+
+ [English-portuguese translation](https://www.kaggle.com/datasets/nageshsingh/englishportuguese-translation).
+
+ The dataset consists of a set of English and Portuguese sentences.
+
+
+## Limitations
+
+Translations are far from perfect. To improve this model, we could:
+
+1. Use a deep stack of recurrent layers for both the encoder and the decoder.
+2. Or, we could use an `LSTM` instead of a `GRU`.
+
+In conclusion, we do not recommend using this model in real-world applications. 
+It was solely developed for academic and educational purposes.
+
+## Cite as 🤗
+
+```latex
 @misc{teenytinycastle,
     doi = {10.5281/zenodo.7112065},
-    url = {https://huggingface.co/AiresPucrs/GRU-eng-por},
+    url = {https://github.com/Nkluge-correa/teeny-tiny_castle},
     author = {Nicholas Kluge Corr{\^e}a},
     title = {Teeny-Tiny Castle},
-    year = {2023},
-    publisher = {HuggingFace},
-    journal = {HuggingFace repository},
+    year = {2024},
+    publisher = {GitHub},
+    journal = {GitHub repository},
 }
 ```
+
 ## License
 The GRU-eng-por is licensed under the Apache License, Version 2.0. See the LICENSE file for more details.