(1) At first, ETM need load word embeddings. You can download the model by Glov2Vec from this url http://nlp.stanford.edu/projects/glove/.
(2) Input the two files (data and word list) using the following form.
Data:
1 3 1 3 4 2 3 4 1
where each line is a document, and the number is the index of the words
Word list: each line is a word.