Update README.md

README.md

@@ -51,13 +51,15 @@ The models have been trained using some of the preprocessing steps from [cc_net]
 * KenLM: `pip install https://github.com/kpu/kenlm/archive/master.zip`
 * SentencePiece: `pip install sentencepiece`
 
+
 # Example:
 ```
-from model import KenlmModel
+from model import KenlmModel # KenlmModel comes from the model.py file at the root of this hf directory.
 
 
 # Load model trained on English wikipedia
-model = KenlmModel.from_pretrained("wikipedia", "en")
+model = KenlmModel.from_pretrained("wikipedia", "en") # You need to download the following files first: en.arpa.bin, en.sp.model and en.sp.vocab from the hf repo
+
 
 # Get perplexity
 model.get_perplexity("I am very perplexed")