Source code for my attempts: https://github.com/Tzeny/udacity-deep-learning/blob/master/5_word2vec.ipynb
In the git repository above I’ve also included the embeddings matrix resulted from the trained in the form of a .pickle file.
Some examples from the skip-gram training:
Nearest to one: two, four, seven, eight, three, six, five, nine,
Nearest to man: person, woman, boy, chanute, glasgow, programmatical, trudeau, revenge,
Nearest to concept: form, idea, locational, testimony, result, definition, tordesillas, nisos,