SYMBOL INDEX (96 symbols across 16 files)

FILE: DQN/dqn.py
  class StateProcessor (line 21) | class StateProcessor():
    method __init__ (line 25) | def __init__(self):
    method process (line 35) | def process(self, sess, state):
  class Estimator (line 46) | class Estimator():
    method __init__ (line 52) | def __init__(self, scope="estimator", summaries_dir=None):
    method _build_model (line 65) | def _build_model(self):
    method predict (line 115) | def predict(self, sess, s):
    method update (line 129) | def update(self, sess, s, a, y):
  function copy_model_parameters (line 150) | def copy_model_parameters(sess, estimator1, estimator2):
  function make_epsilon_greedy_policy (line 172) | def make_epsilon_greedy_policy(estimator, nA):
  function deep_q_learning (line 194) | def deep_q_learning(sess,

FILE: PolicyGradient/a3c/estimator_test.py
  function make_env (line 21) | def make_env():
  class PolicyEstimatorTest (line 26) | class PolicyEstimatorTest(tf.test.TestCase):
    method testPredict (line 27) | def testPredict(self):
    method testGradient (line 54) | def testGradient(self):
  class ValueEstimatorTest (line 81) | class ValueEstimatorTest(tf.test.TestCase):
    method testPredict (line 82) | def testPredict(self):
    method testGradient (line 107) | def testGradient(self):

FILE: PolicyGradient/a3c/estimators.py
  function build_shared_network (line 4) | def build_shared_network(X, add_summaries=False):
  class PolicyEstimator (line 36) | class PolicyEstimator():
    method __init__ (line 49) | def __init__(self, num_outputs, reuse=False, trainable=True):
  class ValueEstimator (line 109) | class ValueEstimator():
    method __init__ (line 120) | def __init__(self, reuse=False, trainable=True):

FILE: PolicyGradient/a3c/policy_monitor.py
  class PolicyMonitor (line 25) | class PolicyMonitor(object):
    method __init__ (line 35) | def __init__(self, env, policy_net, summary_writer, saver=None):
    method _policy_net_predict (line 62) | def _policy_net_predict(self, state, sess):
    method eval_once (line 67) | def eval_once(self, sess):
    method continuous_eval (line 100) | def continuous_eval(self, eval_every, sess, coord):

FILE: PolicyGradient/a3c/policy_monitor_test.py
  function make_env (line 24) | def make_env():
  class PolicyMonitorTest (line 29) | class PolicyMonitorTest(tf.test.TestCase):
    method setUp (line 30) | def setUp(self):
    method testEvalOnce (line 41) | def testEvalOnce(self):

FILE: PolicyGradient/a3c/train.py
  function make_env (line 37) | def make_env(wrap=True):

FILE: PolicyGradient/a3c/worker.py
  function make_copy_params_op (line 24) | def make_copy_params_op(v1_list, v2_list):
  function make_train_op (line 39) | def make_train_op(local_estimator, global_estimator):
  class Worker (line 53) | class Worker(object):
    method __init__ (line 67) | def __init__(self, name, env, policy_net, value_net, global_counter, d...
    method run (line 95) | def run(self, sess, coord, t_max):
    method _policy_net_predict (line 118) | def _policy_net_predict(self, state, sess):
    method _value_net_predict (line 123) | def _value_net_predict(self, state, sess):
    method run_n_steps (line 128) | def run_n_steps(self, n, sess):
    method update (line 155) | def update(self, transitions, sess):

FILE: PolicyGradient/a3c/worker_test.py
  function make_env (line 23) | def make_env():
  class WorkerTest (line 28) | class WorkerTest(tf.test.TestCase):
    method setUp (line 29) | def setUp(self):
    method testPolicyNetPredict (line 42) | def testPolicyNetPredict(self):
    method testValueNetPredict (line 59) | def testValueNetPredict(self):
    method testRunNStepsAndUpdate (line 75) | def testRunNStepsAndUpdate(self):

FILE: lib/atari/helpers.py
  class AtariEnvWrapper (line 3) | class AtariEnvWrapper(object):
    method __init__ (line 7) | def __init__(self, env):
    method __getattr__ (line 10) | def __getattr__(self, name):
    method step (line 13) | def step(self, *args, **kwargs):
  function atari_make_initial_state (line 27) | def atari_make_initial_state(state):
  function atari_make_next_state (line 30) | def atari_make_next_state(state, next_state):

FILE: lib/atari/state_processor.py
  class StateProcessor (line 4) | class StateProcessor():
    method __init__ (line 8) | def __init__(self):
    method process (line 18) | def process(self, state, sess=None):

FILE: lib/envs/blackjack.py
  function cmp (line 5) | def cmp(a, b):
  function draw_card (line 12) | def draw_card(np_random):
  function draw_hand (line 16) | def draw_hand(np_random):
  function usable_ace (line 20) | def usable_ace(hand):  # Does this hand have a usable ace?
  function sum_hand (line 24) | def sum_hand(hand):  # Return current hand total
  function is_bust (line 30) | def is_bust(hand):  # Is this hand a bust?
  function score (line 34) | def score(hand):  # What is the score of this hand (0 if bust)
  function is_natural (line 38) | def is_natural(hand):  # Is this hand a natural blackjack?
  class BlackjackEnv (line 42) | class BlackjackEnv(gym.Env):
    method __init__ (line 67) | def __init__(self, natural=False):
    method reset (line 82) | def reset(self):
    method step (line 85) | def step(self, action):
    method _seed (line 88) | def _seed(self, seed=None):
    method _step (line 92) | def _step(self, action):
    method _get_obs (line 111) | def _get_obs(self):
    method _reset (line 114) | def _reset(self):

FILE: lib/envs/cliff_walking.py
  class CliffWalkingEnv (line 12) | class CliffWalkingEnv(discrete.DiscreteEnv):
    method _limit_coordinates (line 16) | def _limit_coordinates(self, coord):
    method _calculate_transition_prob (line 23) | def _calculate_transition_prob(self, current, delta):
    method __init__ (line 31) | def __init__(self):
    method render (line 57) | def render(self, mode='human', close=False):
    method _render (line 60) | def _render(self, mode='human', close=False):

FILE: lib/envs/discrete.py
  class DiscreteEnv (line 7) | class DiscreteEnv(Env):
    method __init__ (line 23) | def __init__(self, nS, nA, P, isd):
    method seed (line 36) | def seed(self, seed=None):
    method reset (line 40) | def reset(self):
    method step (line 45) | def step(self, a):

FILE: lib/envs/gridworld.py
  class GridworldEnv (line 12) | class GridworldEnv(discrete.DiscreteEnv):
    method __init__ (line 34) | def __init__(self, shape=[4,4]):
    method _render (line 88) | def _render(self, mode='human', close=False):

FILE: lib/envs/windy_gridworld.py
  class WindyGridworldEnv (line 13) | class WindyGridworldEnv(discrete.DiscreteEnv):
    method _limit_coordinates (line 17) | def _limit_coordinates(self, coord):
    method _calculate_transition_prob (line 24) | def _calculate_transition_prob(self, current, delta, winds):
    method __init__ (line 31) | def __init__(self):
    method render (line 58) | def render(self, mode='human', close=False):
    method _render (line 61) | def _render(self, mode='human', close=False):

FILE: lib/plotting.py
  function plot_cost_to_go_mountain_car (line 10) | def plot_cost_to_go_mountain_car(env, estimator, num_tiles=20):
  function plot_value_function (line 28) | def plot_value_function(V, title="Value Function"):
  function plot_episode_stats (line 63) | def plot_episode_stats(stats, smoothing_window=10, noshow=False):