Loading

IIT-M RL-ASSIGNMENT-2-TAXI

Solution for submission 132313

A detailed solution for submission 132313 submitted for challenge IIT-M RL-ASSIGNMENT-2-TAXI

p_milind_venkateswara_varma_ee17b024

What is the notebook about?

Problem - Taxi Environment Algorithms

This problem deals with a taxi environment and stochastic actions. The tasks you have to do are:

  • Implement Policy Iteration
  • Implement Modified Policy Iteration
  • Implement Value Iteration
  • Implement Gauss Seidel Value Iteration
  • Visualize the results
  • Explain the results

How to use this notebook? 📝

  • This is a shared template and any edits you make here will not be saved.You should make a copy in your own drive. Click the "File" menu (top-left), then "Save a Copy in Drive". You will be working in your copy however you like.

  • Update the config parameters. You can define the common variables here

Variable Description
AICROWD_DATASET_PATH Path to the file containing test data. This should be an absolute path.
AICROWD_RESULTS_DIR Path to write the output to.
AICROWD_ASSETS_DIR In case your notebook needs additional files (like model weights, etc.,), you can add them to a directory and specify the path to the directory here (please specify relative path). The contents of this directory will be sent to AIcrowd for evaluation.
AICROWD_API_KEY In order to submit your code to AIcrowd, you need to provide your account's API key. This key is available at https://www.aicrowd.com/participants/me

Setup AIcrowd Utilities 🛠

We use this to bundle the files for submission and create a submission on AIcrowd. Do not edit this block.

In [43]:
!pip install aicrowd-cli > /dev/null

AIcrowd Runtime Configuration 🧷

Get login API key from https://www.aicrowd.com/participants/me

In [44]:
import os

AICROWD_DATASET_PATH = os.getenv("DATASET_PATH", os.getcwd()+"/13d77bb0-b325-4e95-a03b-833eb6694acd_a2_taxi_inputs.zip")
AICROWD_RESULTS_DIR = os.getenv("OUTPUTS_DIR", "results")
In [45]:

API Key valid
Saved API Key successfully!
13d77bb0-b325-4e95-a03b-833eb6694acd_a2_taxi_inputs.zip: 100% 31.2k/31.2k [00:00<00:00, 336kB/s]
In [46]:
!unzip $AICROWD_DATASET_PATH
Archive:  /content/13d77bb0-b325-4e95-a03b-833eb6694acd_a2_taxi_inputs.zip
replace inputs/inputs_base.npy? [y]es, [n]o, [A]ll, [N]one, [r]ename: n
replace inputs/inputs_1.npy? [y]es, [n]o, [A]ll, [N]one, [r]ename: n
replace inputs/inputs_0.npy? [y]es, [n]o, [A]ll, [N]one, [r]ename: n
replace inputs/inputs_2.npy? [y]es, [n]o, [A]ll, [N]one, [r]ename: n
replace targets/targets_2.npy? [y]es, [n]o, [A]ll, [N]one, [r]ename: n
replace targets/targets_0.npy? [y]es, [n]o, [A]ll, [N]one, [r]ename: n
replace targets/targets_1.npy? [y]es, [n]o, [A]ll, [N]one, [r]ename: n
replace targets/targets_base.npy? [y]es, [n]o, [A]ll, [N]one, [r]ename: n
In [47]:
DATASET_DIR = 'inputs/'

Taxi Environment

Read the environment to understand the functions, but do not edit anything

In [48]:
import numpy as np

class TaxiEnv_HW2:
    def __init__(self, states, actions, probabilities, rewards, initial_policy):
        self.possible_states = states
        self._possible_actions = {st: ac for st, ac in zip(states, actions)}
        self._ride_probabilities = {st: pr for st, pr in zip(states, probabilities)}
        self._ride_rewards = {st: rw for st, rw in zip(states, rewards)}
        self.initial_policy = initial_policy
        self._verify()

    def _check_state(self, state):
        assert state in self.possible_states, "State %s is not a valid state" % state

    def _verify(self):
        """ 
        Verify that data conditions are met:
        Number of actions matches shape of next state and actions
        Every probability distribution adds up to 1 
        """
        ns = len(self.possible_states)
        for state in self.possible_states:
            ac = self._possible_actions[state]
            na = len(ac)

            rp = self._ride_probabilities[state]
            assert np.all(rp.shape == (na, ns)), "Probabilities shape mismatch"
        
            rr = self._ride_rewards[state]
            assert np.all(rr.shape == (na, ns)), "Rewards shape mismatch"

            assert np.allclose(rp.sum(axis=1), 1), "Probabilities don't add up to 1"

    def possible_actions(self, state):
        """ Return all possible actions from a given state """
        self._check_state(state)
        return self._possible_actions[state]

    def ride_probabilities(self, state, action):
        """ 
        Returns all possible ride probabilities from a state for a given action
        For every action a list with the returned with values in the same order as self.possible_states
        """
        actions = self.possible_actions(state)
        ac_idx = actions.index(action)
        return self._ride_probabilities[state][ac_idx]

    def ride_rewards(self, state, action):
        actions = self.possible_actions(state)
        ac_idx = actions.index(action)
        return self._ride_rewards[state][ac_idx]

Example of Environment usage

In [49]:
def check_taxienv():
    # These are the values as used in the pdf, but they may be changed during submission, so do not hardcode anything

    states = ['A', 'B', 'C']

    actions = [['1','2','3'], ['1','2'], ['1','2','3']]

    probs = [np.array([[1/2,  1/4,  1/4],
                    [1/16, 3/4,  3/16],
                    [1/4,  1/8,  5/8]]),

            np.array([[1/2,   0,     1/2],
                    [1/16,  7/8,  1/16]]),

            np.array([[1/4,  1/4,  1/2],
                    [1/8,  3/4,  1/8],
                    [3/4,  1/16, 3/16]]),]

    rewards = [np.array([[10,  4,  8],
                        [ 8,  2,  4],
                        [ 4,  6,  4]]),

            np.array([[14,  0, 18],
                        [ 8, 16,  8]]),

            np.array([[10,  2,  8],
                        [6,   4,  2],
                        [4,   0,  8]]),]
    initial_policy = {'A': '1', 'B': '1', 'C': '1'}

    env = TaxiEnv_HW2(states, actions, probs, rewards, initial_policy)
    print("All possible states", env.possible_states)
    print("All possible actions from state B", env.possible_actions('B'))
    print("Ride probabilities from state A with action 2", env.ride_probabilities('A', '2'))
    print("Ride rewards from state C with action 3", env.ride_rewards('C', '3'))

    base_kwargs = {"states": states, "actions": actions, 
                "probabilities": probs, "rewards": rewards,
                "initial_policy": initial_policy}
    return base_kwargs

base_kwargs = check_taxienv()
env = TaxiEnv_HW2(**base_kwargs)
All possible states ['A', 'B', 'C']
All possible actions from state B ['1', '2']
Ride probabilities from state A with action 2 [0.0625 0.75   0.1875]
Ride rewards from state C with action 3 [4 0 8]

Task 1 - Policy Iteration

Run policy iteration on the environment and generate the policy and expected reward

In [50]:
# 1.1 Policy Iteration
def policy_iteration(taxienv, gamma):
    # A list of all the states
    states = taxienv.possible_states
    # Initial values
    values = {s: 0 for s in states}

    # This is a dictionary of states to policies -> e.g {'A': '1', 'B': '2', 'C': '1'}
    policy = taxienv.initial_policy.copy()

    ## Begin code here

    # Hints - 
    # Do not hardcode anything
    # Only the final result is required for the results
    # Put any extra data in "extra_info" dictonary for any plots etc
    # Use the helper functions taxienv.ride_rewards, taxienv.ride_probabilities,  taxienv.possible_actions
    # For terminating condition use the condition exactly mentioned in the pdf

    number_of_ite = 0
    ite = {}
    done = 0
    while done != 1:
      # get Jpi(s) for policy at hand
      delta = 1
      number_of_ite = 0
      while delta >= 1e-8:
        delta = 0
        for s in states:
          j = values[s]

          transtition_probs = taxienv.ride_probabilities(s, policy[s])
          rewards           = taxienv.ride_rewards(s, policy[s])
          Tpi = 0
          for i in range(len(transtition_probs)):
            Tpi += transtition_probs[i]*(rewards[i]+gamma*values[states[i]])
          values[s] = Tpi

          delta = max(delta, abs(j-values[s]))
        number_of_ite += 1 

      # get better policy from Jpi(s) and policy at hand
      done = 1
      for s in states:
        b = policy[s]
        actions = taxienv.possible_actions(s)
        Ta = {}
        for action in actions:
          transtition_probs = taxienv.ride_probabilities(s, action)
          rewards           = taxienv.ride_rewards(s, action)
          Tpi = 0
          for i in range(len(transtition_probs)):
            Tpi += transtition_probs[i]*(rewards[i]+gamma*values[states[i]])
          Ta[action] = Tpi
        
        policy[s] = max(Ta, key=lambda ac: Ta[ac])

        if b != policy[s]:
          done = 0
    
    # Put your extra information needed for plots etc in this dictionary
    extra_info = {"gamma": gamma}

    ## Do not edit below this line

    # Final results
    return {"Expected Reward": values, "Policy": policy}, extra_info

Task 2 - Policy Iteration for multiple values of gamma

Ideally this code should run as is

In [51]:
# 1.2 Policy Iteration with different values of gamma
def run_policy_iteration(env):
    gamma_values = np.arange(5, 100, 5)/100
    results, extra_info = {}, {}
    for gamma in gamma_values:
        results[gamma], extra_info[gamma] = policy_iteration(env, gamma)
    return results, extra_info

results, extra_info = run_policy_iteration(env)

# gamma_values = np.arange(5, 100, 5)/100
# for gamma in gamma_values:
#   print(gamma)
#   print(results[gamma]["Expected Reward"])
#   print(results[gamma]["Policy"])

Task 3 - Modifed Policy Iteration

Implement modified policy iteration (where Value iteration is done for fixed m number of steps)

In [52]:
# 1.3 Modified Policy Iteration
def modified_policy_iteration(taxienv, gamma, m):
    # A list of all the states
    states = taxienv.possible_states
    # Initial values
    values = {s: 0 for s in states}

    # This is a dictionary of states to policies -> e.g {'A': '1', 'B': '2', 'C': '1'}
    policy = taxienv.initial_policy.copy()

    ## Begin code here

    # Hints - 
    # Do not hardcode anything
    # Only the final result is required for the results
    # Put any extra data in "extra_info" dictonary for any plots etc
    # Use the helper functions taxienv.ride_rewards, taxienv.ride_probabilities,  taxienv.possible_actions
    # For terminating condition use the condition exactly mentioned in the pdf
    TJ = {s: 0 for s in states}
    ite = 0

    done = 0
    while done != 1:
      ite += 1
      # get Jpi(s) for policy at hand
      for i in range(m):
        for s in states:
          TJ[s] = 0
          transtition_probs = taxienv.ride_probabilities(s, policy[s])
          rewards           = taxienv.ride_rewards(s, policy[s])
          for i in range(len(transtition_probs)):
            TJ[s] += transtition_probs[i]*(rewards[i]+gamma*values[states[i]])      

        for s in states:
          values[s] = TJ[s]
      
      # get better policy from Jpi(s) and policy at hand
      done = 1
      for s in states:
        b = policy[s]
        actions = taxienv.possible_actions(s)
        Ta = {}
        for action in actions:
          transtition_probs = taxienv.ride_probabilities(s, action)
          rewards           = taxienv.ride_rewards(s, action)
          Tpi = 0
          for i in range(len(transtition_probs)):
            Tpi += transtition_probs[i]*(rewards[i]+gamma*values[states[i]])
          Ta[action] = Tpi
        
        policy[s] = max(Ta, key=lambda ac: Ta[ac])

        if b != policy[s]:
          done = 0
    
    # Put your extra information needed for plots etc in this dictionary
    extra_info = {"number_of_iterations": ite}

    ## Do not edit below this line


    # Final results
    return {"Expected Reward": values, "Policy": policy}, extra_info

Task 4 Modified policy iteration for multiple values of m

Ideally this code should run as is

In [53]:
def run_modified_policy_iteration(env):
    m_values = np.arange(1, 15)
    gamma = 0.9
    results, extra_info = {}, {}
    for m in m_values:
        results[m], extra_info[m] = modified_policy_iteration(env, gamma, m)
    return results, extra_info

results, extra_info = run_modified_policy_iteration(env)

# m_values = np.arange(1, 15)
# for m in m_values:
#   print(m)
#   print(results[m]["Expected Reward"])
#   print(results[m]["Policy"])
#   print(extra_info[m]["number_of_iterations"])

Task 5 Value Iteration

Implement value iteration and find the policy and expected rewards

In [54]:
# 1.4 Value Iteration
def value_iteration(taxienv, gamma):
    # A list of all the states
    states = taxienv.possible_states
    # Initial values
    values = {s: 0 for s in states}

    # This is a dictionary of states to policies -> e.g {'A': '1', 'B': '2', 'C': '1'}
    policy = taxienv.initial_policy.copy()

    ## Begin code here

    # Hints - 
    # Do not hardcode anything
    # Only the final result is required for the results
    # Put any extra data in "extra_info" dictonary for any plots etc
    # Use the helper functions taxienv.ride_rewards, taxienv.ride_probabilities,  taxienv.possible_actions
    # For terminating condition use the condition exactly mentioned in the pdf
    H = {s: 0 for s in states}
    ite = 0
    delta = 1
    while delta >= 1e-8:
      ite += 1
      delta = 0
      for s in states:
        actions = taxienv.possible_actions(s)
        Hs = {action: 0 for action in actions}
        for action in actions:
          transtition_probs = taxienv.ride_probabilities(s, action)
          rewards           = taxienv.ride_rewards(s, action)
          T = 0
          for i in range(len(transtition_probs)):
            T += transtition_probs[i]*(rewards[i]+gamma*values[states[i]])
          Hs[action] = T

        policy[s] = max(Hs, key=lambda ac: Hs[ac])
        H[s] = max(Hs.values())        
        delta = max(delta, abs(H[s]-values[s]))

      for s in states:
        values[s] = H[s]


    # Put your extra information needed for plots etc in this dictionary
    extra_info = {"number_of_iterations": ite}

    ## Do not edit below this line

    # Final results
    return {"Expected Reward": values, "Policy": policy}, extra_info

Task 6 Value Iteration with multiple values of gamma

Ideally this code should run as is

In [55]:
def run_value_iteration(env):
    gamma_values = np.arange(5, 100, 5)/100
    results = {}
    results, extra_info = {}, {}
    for gamma in gamma_values:
        results[gamma], extra_info[gamma] = value_iteration(env, gamma)
    return results, extra_info
  
results, extra_info = run_value_iteration(env)
# gamma_values = np.arange(5, 100, 5)/100
# gamma = 0.9
# print(gamma)
# print(results[gamma]["Expected Reward"])
# print(results[gamma]["Policy"])
# print(extra_info[gamma]["number_of_iterations"])

Task 7 Gauss Seidel Value Iteration

Implement Gauss Seidel Value Iteration

In [56]:
# 1.4 Gauss Seidel Value Iteration
def gauss_seidel_value_iteration(taxienv, gamma):
    # A list of all the states
    # For Gauss Seidel Value Iteration - iterate through the values in the same order
    states = taxienv.possible_states

    # Initial values
    values = {s: 0 for s in states}

    # This is a dictionary of states to policies -> e.g {'A': '1', 'B': '2', 'C': '1'}
    policy = taxienv.initial_policy.copy()

    # Hints - 
    # Do not hardcode anything
    # For Gauss Seidel Value Iteration - iterate through the values in the same order as taxienv.possible_states
    # Only the final result is required for the results
    # Put any extra data in "extra_info" dictonary for any plots etc
    # Use the helper functions taxienv.ride_rewards, taxienv.ride_probabilities,  taxienv.possible_actions
    # For terminating condition use the condition exactly mentioned in the pdf

    ## Begin code here
    delta = 1
    ite = 0
    while delta >= 1e-8:
      ite += 1
      delta = 0
      for s in states:
        j = values[s]
        actions = taxienv.possible_actions(s)
        Hs = {action: 0 for action in actions}
        for action in actions:
          transtition_probs = taxienv.ride_probabilities(s, action)
          rewards           = taxienv.ride_rewards(s, action)
          T = 0
          for i in range(len(transtition_probs)):
            T += transtition_probs[i]*(rewards[i]+gamma*values[states[i]])
          Hs[action] = T

        policy[s] = max(Hs, key=lambda ac: Hs[ac])
        values[s] = max(Hs.values())        
        delta = max(delta, abs(j-values[s]))

    # Put your extra information needed for plots etc in this dictionary
    extra_info = {"number_of_iterations": ite}

    ## Do not edit below this line

    # Final results
    return {"Expected Reward": values, "Policy": policy}, extra_info

Task 8 Gauss Seidel Value Iteration with multiple values of gamma

Ideally this code should run as is

In [57]:
def run_gauss_seidel_value_iteration(env):
    gamma_values = np.arange(5, 100, 5)/100
    results = {}
    results, extra_info = {}, {}
    for gamma in gamma_values:
        results[gamma], extra_info[gamma] = gauss_seidel_value_iteration(env, gamma)
    return results, extra_info

results, extra_info = run_gauss_seidel_value_iteration(env)
# gamma_values = np.arange(5, 100, 5)/100
# gamma = 0.9
# # for gamma in gamma_values:
# print(gamma)
# print(results[gamma]["Expected Reward"])
# print(results[gamma]["Policy"])
# print(extra_info[gamma]["number_of_iterations"])

Generate Results ✅

In [58]:
# Do not edit this cell
def get_results(kwargs):

    taxienv = TaxiEnv_HW2(**kwargs)

    policy_iteration_results = run_policy_iteration(taxienv)[0]
    modified_policy_iteration_results = run_modified_policy_iteration(taxienv)[0]
    value_iteration_results = run_value_iteration(taxienv)[0]
    gs_vi_results = run_gauss_seidel_value_iteration(taxienv)[0]

    final_results = {}
    final_results["policy_iteration"] = policy_iteration_results
    final_results["modifed_policy_iteration"] = modified_policy_iteration_results
    final_results["value_iteration"] = value_iteration_results
    final_results["gauss_seidel_iteration"] = gs_vi_results

    return final_results
In [ ]:
# Do not edit this cell, generate results with it as is
if not os.path.exists(AICROWD_RESULTS_DIR):
    os.mkdir(AICROWD_RESULTS_DIR)

for params_file in os.listdir(DATASET_DIR):
  kwargs = np.load(os.path.join(DATASET_DIR, params_file), allow_pickle=True).item()
  results = get_results(kwargs)
  idx = params_file.split('_')[-1][:-4]
  np.save(os.path.join(AICROWD_RESULTS_DIR, 'results_' + idx), results)

Check your local score

This score is not your final score, and it doesn't use the marks weightages. This is only for your reference of how arrays are matched and with what tolerance.

In [ ]:
# Check your score on the given test cases (There are more private test cases not provided)
target_folder = 'targets'
result_folder = AICROWD_RESULTS_DIR

def check_algo_match(results, targets):
    param_matches = []
    for k in results:
        param_results = results[k]
        param_targets = targets[k]
        policy_match = param_results['Policy'] == param_targets['Policy']
        rv = [v for k, v in param_results['Expected Reward'].items()]
        tv = [v for k, v in param_targets['Expected Reward'].items()]
        rewards_match = np.allclose(rv, tv, rtol=3)
        equal = rewards_match and policy_match
        param_matches.append(equal)
    return np.mean(param_matches)

def check_score(target_folder, result_folder):
    match = []
    for out_file in os.listdir(result_folder):
        res_file = os.path.join(result_folder, out_file)
        results = np.load(res_file, allow_pickle=True).item()
        idx = out_file.split('_')[-1][:-4]  # Extract the file number
        target_file = os.path.join(target_folder, f"targets_{idx}.npy")
        targets = np.load(target_file, allow_pickle=True).item()
        algo_match = []
        for k in targets:
            algo_results = results[k]
            algo_targets = targets[k]
            algo_match.append(check_algo_match(algo_results, algo_targets))
        match.append(np.mean(algo_match))
    return np.mean(match)

if os.path.exists(target_folder):
    print("Shared data Score (normalized to 1):", check_score(target_folder, result_folder))

Visualize results of Policy Iteration with multiple values of gamma

Add code to visualize the results

In [ ]:
## Visualize policy iteration with multiple values of gamma

# Running modified policy iteration to get results
def run_policy_iteration(env):
    gamma_values = np.arange(5, 100, 5)/100
    results, extra_info = {}, {}
    for gamma in gamma_values:
        results[gamma], extra_info[gamma] = policy_iteration(env, gamma)
    return results, extra_info

results, extra_info = run_policy_iteration(env)

# presenting results
gamma_values = np.arange(5, 100, 5)/100
for gamma in gamma_values:
  print("For gamme value: " + "{:.2f}".format(gamma) + " ", end="")
  print(results[gamma]["Policy"])

Subjective questions

1.a How are values of $\gamma$ affecting results of policy iteration

As gamma value increases the policy iteration starts choosing the policy which is better suited for multiple trips, the policy being "go to taxi stand" for every city. When gamma is lower the policy iteration chooses the most immediate profitable action, which is "cruise the streets for a passenger" for every city.

1.b For modified policy itetaration, do you find any improvement if you choose m=10.

For m=5 the results are:
J : {'A': 89, 'B': 103, 'C': 90} and
pi: {'A': '2', 'B': '2', 'C': '2'}

For m=10 the results are 10
{'A': 113, 'B': 126, 'C': 114}
{'A': '2', 'B': '2', 'C': '2'}

The final policy in both instances is the same but the cost values calculated for m=10 are more accurate

1.c Compare and contrast the behavior of Value Iteration and Gauss Seidel Value Iteraton

For gamma = 0.9 both value iteration and gauss seidel value iteration give the same final cost values and policy.
But value iteration takes 201 iterations while gauss seidel takes 179 iterations.

Submit to AIcrowd 🚀

In [ ]:
!DATASET_PATH=$AICROWD_DATASET_PATH aicrowd notebook submit -c iit-m-rl-assignment-2-taxi -a assets
922

Comments

You must login before you can post a comment.

Execute