In [3]:
import numpy as np
import pprint
import sys
if "../" not in sys.path:
  sys.path.append("../") 
from lib.envs.gridworld import GridworldEnv

In [4]:
pp = pprint.PrettyPrinter(indent=2)
env = GridworldEnv()

In [5]:
def value_iteration(env, theta=0.0001, discount_factor=1.0):
    """
    Value Iteration Algorithm.
    
    Args:
        env: OpenAI env. env.P represents the transition probabilities of the environment.
            env.P[s][a] is a list of transition tuples (prob, next_state, reward, done).
            env.nS is a number of states in the environment. 
            env.nA is a number of actions in the environment.
        theta: We stop evaluation once our value function change is less than theta for all states.
        discount_factor: Gamma discount factor.
        
    Returns:
        A tuple (policy, V) of the optimal policy and the optimal value function.        
    """
    

    V = np.zeros(env.nS)
    policy = np.zeros([env.nS, env.nA])
    
    # Implement!
    return policy, V

In [6]:
policy, v = value_iteration(env)

print("Policy Probability Distribution:")
print(policy)
print("")

print("Reshaped Grid Policy (0=up, 1=right, 2=down, 3=left):")
print(np.reshape(np.argmax(policy, axis=1), env.shape))
print("")

print("Value Function:")
print(v)
print("")

print("Reshaped Grid Value Function:")
print(v.reshape(env.shape))
print("")


Policy Probability Distribution:
[[ 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.]]

Reshaped Grid Policy (0=up, 1=right, 2=down, 3=left):
[[0 0 0 0]
 [0 0 0 0]
 [0 0 0 0]
 [0 0 0 0]]

Value Function:
[ 0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.]

Reshaped Grid Value Function:
[[ 0.  0.  0.  0.]
 [ 0.  0.  0.  0.]
 [ 0.  0.  0.  0.]
 [ 0.  0.  0.  0.]]


In [7]:
# Test the value function
expected_v = np.array([ 0, -1, -2, -3, -1, -2, -3, -2, -2, -3, -2, -1, -3, -2, -1,  0])
np.testing.assert_array_almost_equal(v, expected_v, decimal=2)


---------------------------------------------------------------------------
AssertionError                            Traceback (most recent call last)
<ipython-input-7-55581f8eb5c9> in <module>()
      1 # Test the value function
      2 expected_v = np.array([ 0, -1, -2, -3, -1, -2, -3, -2, -2, -3, -2, -1, -3, -2, -1,  0])
----> 3 np.testing.assert_array_almost_equal(v, expected_v, decimal=2)

/Users/dennybritz/venvs/tf/lib/python3.5/site-packages/numpy/testing/utils.py in assert_array_almost_equal(x, y, decimal, err_msg, verbose)
    914     assert_array_compare(compare, x, y, err_msg=err_msg, verbose=verbose,
    915              header=('Arrays are not almost equal to %d decimals' % decimal),
--> 916              precision=decimal)
    917 
    918 

/Users/dennybritz/venvs/tf/lib/python3.5/site-packages/numpy/testing/utils.py in assert_array_compare(comparison, x, y, err_msg, verbose, header, precision)
    735                                 names=('x', 'y'), precision=precision)
    736             if not cond:
--> 737                 raise AssertionError(msg)
    738     except ValueError:
    739         import traceback

AssertionError: 
Arrays are not almost equal to 2 decimals

(mismatch 87.5%)
 x: array([ 0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,  0.,
        0.,  0.,  0.])
 y: array([ 0, -1, -2, -3, -1, -2, -3, -2, -2, -3, -2, -1, -3, -2, -1,  0])