IIT-M RL-ASSIGNMENT-2-TAXI
Solution for submission 132341
A detailed solution for submission 132341 submitted for challenge IIT-M RL-ASSIGNMENT-2-TAXI
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 |
- Installing packages. Please use the Install packages 🗃 section to install the packages
Setup AIcrowd Utilities 🛠¶
We use this to bundle the files for submission and create a submission on AIcrowd. Do not edit this block.
!pip install aicrowd-cli > /dev/null
AIcrowd Runtime Configuration 🧷¶
Get login API key from https://www.aicrowd.com/participants/me
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")
!unzip $AICROWD_DATASET_PATH
DATASET_DIR = 'inputs/'
Taxi Environment¶
Read the environment to understand the functions, but do not edit anything
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¶
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)
Task 1 - Policy Iteration¶
Run policy iteration on the environment and generate the policy and expected reward
# 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
flag = False
while not flag:
#policy evaluation step
while True:
delta =0
values_old =values.copy()
for s in states:
a=policy[s]
J= [ i*gamma for i in list(values_old.values())]
values[s] = sum(taxienv.ride_probabilities(s,a)* ( taxienv.ride_rewards(s,a) + J ))
delta = max(delta, abs(values[s] - values_old[s]))
if delta < 1e-8:
break
#policy improvement step
flag =True
for s in states:
actions = taxienv.possible_actions(s)
rewards = {a:0 for a in actions}
for a in actions:
J= [ i*gamma for i in list(values.values()) ]
rewards[a] = sum(taxienv.ride_probabilities(s, a)* ( taxienv.ride_rewards(s, a) + J ))
action=max(rewards, key=rewards.get)
if policy[s] != action:
policy[s]=action
flag= False
# 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
# Put your extra information needed for plots etc in this dictionary
extra_info = {}
## 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
# 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)
Task 3 - Modifed Policy Iteration¶
Implement modified policy iteration (where Value iteration is done for fixed m number of steps)
# 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
flag = False
while not flag:
#policy evaluation step
for i in range(m):
values_old =values.copy()
for s in states:
a=policy[s]
J= [ i*gamma for i in list(values_old.values())]
values[s] = sum(taxienv.ride_probabilities(s,a)* ( taxienv.ride_rewards(s,a) + J ))
#policy improvement step
flag =True
for s in states:
actions = taxienv.possible_actions(s)
rewards = {a:0 for a in actions}
for a in actions:
J= [ i*gamma for i in list(values.values()) ]
reward = sum(taxienv.ride_probabilities(s, a)* ( taxienv.ride_rewards(s, a) + J ))
rewards[a]=reward
action=max(rewards, key=rewards.get)
if policy[s] != action:
policy[s]=action
flag= False
# 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
# Put your extra information needed for plots etc in this dictionary
extra_info = {}
## 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
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)
Task 5 Value Iteration¶
Implement value iteration and find the policy and expected rewards
# 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
iter=0
delta_values=[]
while True:
iter=iter+1
delta=0
values_old=values.copy()
for s in states:
actions=taxienv.possible_actions(s)
rewards= {a:0 for a in actions}
for a in actions:
J= [ i*gamma for i in list(values_old.values()) ]
rewards[a] = sum(taxienv.ride_probabilities(s, a)* ( taxienv.ride_rewards(s, a) + J ))
action=max(rewards,key=rewards.get)
values[s]=rewards[action]
policy[s]=action
delta=max(delta,abs(values[s]-values_old[s]))
delta_values.append(delta)
if delta < 1e-8:
break
# 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
# Put your extra information needed for plots etc in this dictionary
extra_info = {"iter":iter}
## 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
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)
Task 7 Gauss Seidel Value Iteration¶
Implement Gauss Seidel Value Iteration
# 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
iter=0
delta_values=[]
while True:
iter=iter+1
delta=0
values_old=values.copy()
for s in states:
actions=taxienv.possible_actions(s)
rewards= {a:0 for a in actions}
for a in actions:
J= [ i*gamma for i in list(values.values()) ]
rewards[a] = sum(taxienv.ride_probabilities(s, a)* ( taxienv.ride_rewards(s, a) + J ))
action=max(rewards,key=rewards.get)
values[s]=rewards[action]
policy[s]=action
delta=max(delta,abs(values[s]-values_old[s]))
delta_values.append(delta)
if delta <1e-8:
break
# Put your extra information needed for plots etc in this dictionary
extra_info = {"iter":iter}
## 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
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)
Generate Results ✅¶
# 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
# 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.
# 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))
gamma = 0.9
results, extra_info = policy_iteration(env, gamma)
reward = results["Expected Reward"]
policy = results["Policy"]
print("Optimal Policy",policy)
print("Expected Rewards",reward)
Optimal Policy is to Always take action 2 i.e Go to the nearest taxi stand and wait in line.¶
1.2: Run policy iteration for discount factors γ ranging from 0 to 0.95 with intervals of 0.05 and display the results.¶
results, extra_info = run_policy_iteration(env)
import matplotlib.pyplot as plt
plt.figure(figsize=(10, 5), dpi=100)
plt.title("Expected Reward Vs Gamma Values Plot")
plt.xlabel('Gamma values')
plt.ylabel('Expected rewards')
plt.xticks(list(results.keys()))
cities = ['A','B','C']
for city in cities:
plt.plot(list(results.keys()),[i['Expected Reward'][city] for i in results.values()],marker='o',linewidth=0.5, markersize=3)
plt.legend(["City_A","City_B","City_C"])
plt.show()
##printing all rewards and policy wrt to gamma
print("Expected Reward")
for gamma in results.keys():
rewards = results[gamma]['Expected Reward']
print("Gamma",gamma,rewards)
print("\nPolicy")
for gamma in results.keys():
policy = results[gamma]['Policy']
print("Gamma",gamma,policy)
1.3: Find an optimal policy using modified policy iteration(Algorithm 4) starting with a policy that will always cruise independent of the town, and a zero value vector. Let γ = 0.9 and m = 5¶
##simply just running modified PI with given values
gamma = 0.9
m = 5
results, extra_info = modified_policy_iteration(env, gamma, m)
rewards = results["Expected Reward"]
policy = results["Policy"]
print("Optimal Policy ",policy)
print("Expected Rewards",rewards)
For Modified PI , Optimal Policy is to Always take action 2 i.e Go to the nearest taxi stand and wait in line¶
1.4: Find optimal values using value iteration(Algorithm 1) starting with a zero vector. Let γ = 0.9.¶
gamma = 0.9
results, extra_info = value_iteration(env, gamma)
rewards = results["Expected Reward"]
policy = results["Policy"]
print("Optimal Policy",policy)
print("Expected Rewards",rewards)
For VI , Optimal Policy is to Always take action 2 i.e Go to the nearest taxi stand and wait in line. Expected Rewards are¶
{'A': 121.65347103895924, 'B': 135.30627543932593, 'C': 122.83690299162194}
1.5: Find optimal values using Gauss-Seidel value iteration(Algorithm 2) starting with a zero vector. Let γ = 0.9.¶
gamma = 0.9
results, extra_info = gauss_seidel_value_iteration(env, gamma)
rewards = results["Expected Reward"]
policy = results["Policy"]
print("Optimal Policy",policy)
print("Expected Rewards",rewards)
For Gauss-Seidel VI , Optimal Policy is to Always take action 2 i.e Go to the nearest taxi stand and wait in line. Expected Rewards are¶
{'A': 121.65347104963263, 'B': 135.30627544959796, 'C': 122.83690300915262}
Subjective questions¶
1.a How are values of $\gamma$ affecting results of policy iteration¶
We can see from the graph that higher $\gamma$ gives more Expected reward. This can be reasond as follows
since 1-$\gamma$ is the probabilty our taxi drivers car will break down, the longer the care can run , the more money he can make.
If $\gamma$ was 0 , his car would immediatly break and he cannot make money
If $\gamma$ was 1 , his car will never break and he'll make a treamendous (tending to infinity) amount of money
results, extra_info = run_policy_iteration(env)
import matplotlib.pyplot as plt
plt.figure(figsize=(10, 5), dpi=100)
plt.title("Expected Reward Vs Gamma Values Plot")
plt.xlabel('Gamma values')
plt.ylabel('Expected rewards')
plt.xticks(list(results.keys()))
cities = ['A','B','C']
for city in cities:
plt.plot(list(results.keys()),[i['Expected Reward'][city] for i in results.values()],marker='o',linewidth=0.5, markersize=3)
plt.legend(["City_A","City_B","City_C"])
plt.show()
""" our policy changes as follows
Gamma 0.05 {'A': '1', 'B': '1', 'C': '1'}
Gamma 0.1 {'A': '1', 'B': '1', 'C': '1'}
Gamma 0.15 {'A': '1', 'B': '2', 'C': '1'}
Gamma 0.2 {'A': '1', 'B': '2', 'C': '1'}
Gamma 0.25 {'A': '1', 'B': '2', 'C': '1'}
Gamma 0.3 {'A': '1', 'B': '2', 'C': '1'}
Gamma 0.35 {'A': '1', 'B': '2', 'C': '1'}
Gamma 0.4 {'A': '1', 'B': '2', 'C': '1'}
Gamma 0.45 {'A': '1', 'B': '2', 'C': '1'}
Gamma 0.5 {'A': '1', 'B': '2', 'C': '1'}
Gamma 0.55 {'A': '1', 'B': '2', 'C': '2'}
Gamma 0.6 {'A': '1', 'B': '2', 'C': '2'}
Gamma 0.65 {'A': '1', 'B': '2', 'C': '2'}
Gamma 0.7 {'A': '1', 'B': '2', 'C': '2'}
Gamma 0.75 {'A': '1', 'B': '2', 'C': '2'}
Gamma 0.8 {'A': '2', 'B': '2', 'C': '2'}
Gamma 0.85 {'A': '2', 'B': '2', 'C': '2'}
Gamma 0.9 {'A': '2', 'B': '2', 'C': '2'}
Gamma 0.95 {'A': '2', 'B': '2', 'C': '2'}
we can easily see how it changes from "1" ie cruise the streets to look for a passenger to "2" ie go to nearest taxi line and wait
this can be reasoned as with low gamma driver doesn't have alot of trips he can make and therefore chooses "1" which is the short term
maximizer , while "2" is the long term maximizer which he can choose when the car has low change of breaking down"""
gamma = 0.9
m = 5
results5, extra_info = modified_policy_iteration(env, gamma, m)
rewards5 = results5["Expected Reward"]
policy5 = results5["Policy"]
print("Optimal Policy" ,policy5)
print("Expected Rewards",rewards5)
gamma = 0.9
m = 10
results10, extra_info = modified_policy_iteration(env, gamma, m)
rewards10 = results10["Expected Reward"]
policy10 = results10["Policy"]
print("\nOptimal Policy",policy10)
print("Expected Rewards ",rewards10)
Another thing to consider is how our Expected reward from modified PI is now closer to the expected reward from normal PI. This would indicate that we are getting more accurate values of expected reward with m=10 when compared with m=5 , and the difference here is significant. at higher values of m , say m=15 or 20 it could converge and therefore not show significant improvment , but between 5 and 10 it is showing¶
1.c Compare and contrast the behavior of Value Iteration and Gauss Seidel Value Iteraton¶
Here we see that VI reaches completion in more number of iterations than GS for all values of Gamma. Hence we can say that GS is a better algorithm as it gives us the answer in lower number of iterations.
import matplotlib.pyplot as plt
resultsvi, extra_infovi = run_value_iteration(env)
resultsgs, extra_infogs = run_gauss_seidel_value_iteration(env)
plt.figure(figsize=(10, 5), dpi=100)
plt.xlabel('Gamma')
plt.ylabel('Iterations')
plt.xticks(list(extra_infovi.keys()))
VI_iter= [extra_infovi[gamma]['iter'] for gamma in extra_infovi.keys()]
GS_iter=[extra_infogs[gamma]['iter'] for gamma in extra_infogs.keys()]
plt.plot(list(extra_infovi.keys()), VI_iter, color='cyan', marker='o', linewidth=0.5, markersize=5, label = "VI")
plt.plot(list(extra_infogs.keys()), GS_iter, color='red', marker='o', linewidth=0.5, markersize=5, label = "GS")
plt.legend()
plt.show()
Submit to AIcrowd 🚀¶
!DATASET_PATH=$AICROWD_DATASET_PATH aicrowd notebook submit -c iit-m-rl-assignment-2-taxi -a assets
Content
Comments
You must login before you can post a comment.