Example `lasso_missing.py` is broken in master

$ python lasso_missing.py
Traceback (most recent call last):
  File "lasso_missing.py", line 42, in <module>
    start = pm.find_MAP()
  File "/tmp/pymc3/pymc3/tuning/starting.py", line 168, in find_MAP
    specific_errors)
ValueError: Optimization error: max, logp or dlogp at max have non-finite values. Some values may be outside of distribution support. max: {'sib_mean_log__': array(-0.366512917833829), 'siblings_imp_missing': array([], dtype=int64), 'p_disab_logodds__': array(0.0), 'disability_imp_
missing': array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]), 'p_mother_logodds__': array(0.0), 'mother_imp_missing': array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
       0, 0, 0, 0]), 's_log__': array(1.6094379124341003), 'beta': array([ 0.1,  0.1,  0.1,  0.1,  0.1,  0.1,  0.1])} logp: array(nan) dlogp: array([  8.98253860e+01,  -5.85000000e+01,  -9.50000000e+00,
         5.58700716e+04,   6.49722000e+02,   3.32594000e+02,
         7.25314000e+02,   1.26742000e+02,   3.48175580e+04,
         3.19726000e+02,   2.59158000e+02])Check that 1) you don't have hierarchical parameters, these will lead to points with infinite density. 2) your distribution logp's are properly specified. Specific issues: 
beta.logp bad: nan