My Code (given below) calculated the word embeddings using svd2vec model and converts it into gensim formatted keyvector as:
from svd2vec import svd2vec from gens