Update infersent readme

Author: dkutin

infersent/README.md

@@ -19,13 +19,21 @@ curl -Lo GloVe/glove.840B.300d.zip http://nlp.stanford.edu/data/glove.840B.300d.
 unzip GloVe/glove.840B.300d.zip -d GloVe/
 ```
 
-## Use our sentence encoder
-We provide a simple interface to encode English sentences. **See [**demo.ipynb**](https://github.com/facebookresearch/InferSent/blob/master/demo.ipynb)
-for a practical example.** Get started with the following steps:
+## Download sentence encoder
 
-*0.0) Download our InferSent models (V1 trained with GloVe, V2 trained with fastText)[147MB]:*
+### InferSent models trained with GloVe[147MB]:
 ```bash
 mkdir encoder
 curl -Lo encoder/infersent1.pkl https://dl.fbaipublicfiles.com/infersent/infersent1.pkl
 ```
-Note that infersent1 is trained with GloVe (which have been trained on text preprocessed with the PTB tokenizer) and infersent2 is trained with fastText (which have been trained on text preprocessed with the MOSES tokenizer). The latter also removes the padding of zeros with max-pooling which was inconvenient when embedding sentences outside of their batches.
+
+## Download necessary libraries
+Run the following code below in the terminal to download the libraries:
+- `pip3 install --user numpy`
+- `pip3 install --user torch`
+- `pip3 install --user nltk`
+
+After installing all required packages, we will also need to download and format the sentence encoder & pre-trained model. 
+
+## Run the program
+- To run the Infersent approach, call the `main.py` file inside the `/infersent` directory after set up steps have been completed with `python3 main.py`.