This program utilizes Hidden Markov Models to generate random Shakespeare-esque sonnets.

HMM, Poem Generation, and Visualization Instructions
----------------------------------------------------

Main Files:
    hmm.py (Baum-Welch algorithm)
    hmm_helper.py (Preprocessing, Tokenization, Saving Matrices)
    train.py (Off-the-shelf algorithm example)
    data_vis.py (Data Visualization)
    poem_gen.py (Poem Generation)

HMM Algorithm:
    from hmm.py import HMM
    # EXAMPLE
    h = HMM(2)
    data = [['R', 'W', 'B', 'B']] # One sequence
    h.train(data)
    print h.A
    print h.O
    print h.PI

Train Model and Save:
    from hmm_helper import *
    train_model("Model_name", 20, data)

Visualization:
    from hmm_helper import *
    trans, emiss, wordmap, init = load_model("Model_name")
    count = gen_count_dict()
    clean_print_top(find_best(emiss, wordmap, count))
    clean_print_trans(trans_max(trans))

Poem Generation:
    from poem_gen import *
    # Generate unrhymed, but stressed poems:
    poem1 = gpoem_qac_models("Model_name")
    # Generate ryhmed but unstressed poems:
    poem 2 = gpoem_rev_models("Model_name")