n_rows = 4
n_columns = 4
obstacles = [(1,1), (1,3)]
s_size=1, a_size=4,h_size=32
lr=1e-2
n_training_episodes=1500, max_t=200, gamma=0.99
batches=20
alpha = 0.9

RESULT - 
 12/20 converged
 2/12 converged late
keeping the good cases and getting rid of others (keeping all)