import numpy as np import tensorflow as tf import tensorflow.contrib.layers as c_layers from tensorflow.python.tools import freeze_graph from unityagents import UnityEnvironmentException def create_agent_model(env, lr=1e-4, h_size=128, epsilon=0.2, beta=1e-3, max_step=5e6): """ Takes a Unity environment and model-specific hyperparameters and returns the appropriate PPO agent model for the environment. :param env: a Unity environment. :param lr: Learning rate. :param h_size: Size of hidden layers/ :param epsilon: Value for policy-divergence threshold. :param beta: Strength of entropy regularization. :return: a sub-class of PPOAgent tailored to the environment. """ brain_name = env.brain_names[0] if env.brains[brain_name].action_space_type == "continuous": if env.brains[brain_name].number_observations == 0: return ContinuousControlModel(lr, env.brains[brain_name].state_space_size, env.brains[brain_name].action_space_size, h_size, epsilon, beta, max_step) else: raise UnityEnvironmentException("There is currently no PPO model which supports both a continuous " "action space and camera observations.") if env.brains[brain_name].action_space_type == "discrete": if env.brains[brain_name].number_observations == 0: return DiscreteControlModel(lr, env.brains[brain_name].state_space_size, env.brains[brain_name].action_space_size, h_size, epsilon, beta, max_step) else: brain = env.brains[brain_name] if env.brains[brain_name].state_space_size > 0: print("This brain contains agents with both observations and states. There is currently no PPO model" "which supports this. Defaulting to Vision-based PPO model.") h, w = brain.camera_resolutions[0]['height'], brain.camera_resolutions[0]['height'] return VisualDiscreteControlModel(lr, h, w, env.brains[brain_name].action_space_size, h_size, epsilon, beta, max_step) def save_model(sess, saver, model_path="./", steps=0): """ Saves current model to checkpoint folder. :param sess: Current Tensorflow session. :param model_path: Designated model path. :param steps: Current number of steps in training process. :param saver: Tensorflow saver for session. """ last_checkpoint = model_path + '/model-' + str(steps) + '.cptk' saver.save(sess, last_checkpoint) tf.train.write_graph(sess.graph_def, model_path, 'raw_graph_def.pb', as_text=False) print("Saved Model") def export_graph(model_path, env_name="env", target_nodes="action"): """ Exports latest saved model to .bytes format for Unity embedding. :param model_path: path of model checkpoints. :param env_name: Name of associated Learning Environment. :param target_nodes: Comma separated string of needed output nodes for embedded graph. """ ckpt = tf.train.get_checkpoint_state(model_path) freeze_graph.freeze_graph(input_graph=model_path + '/raw_graph_def.pb', input_binary=True, input_checkpoint=ckpt.model_checkpoint_path, output_node_names=target_nodes, output_graph=model_path + '/' + env_name + '.bytes', clear_devices=True, initializer_nodes="", input_saver="", restore_op_name="save/restore_all", filename_tensor_name="save/Const:0") class PPOModel(object): def __init__(self, probs, old_probs, value, entropy, beta, epsilon, lr, max_step): """ Creates training-specific Tensorflow ops for PPO models. :param probs: Current policy probabilities :param old_probs: Past policy probabilities :param value: Current value estimate :param beta: Entropy regularization strength :param entropy: Current policy entropy :param epsilon: Value for policy-divergence threshold :param lr: Learning rate """ self.returns_holder = tf.placeholder(shape=[None], dtype=tf.float32, name='discounted_rewards') self.advantage = tf.placeholder(shape=[None, 1], dtype=tf.float32, name='advantages') r_theta = probs / old_probs p_opt_a = r_theta * self.advantage p_opt_b = tf.clip_by_value(r_theta, 1 - epsilon, 1 + epsilon) * self.advantage self.policy_loss = -tf.reduce_mean(tf.minimum(p_opt_a, p_opt_b)) self.value_loss = tf.reduce_mean(tf.squared_difference(self.returns_holder, tf.reduce_sum(value, axis=1))) self.loss = self.policy_loss + self.value_loss - beta * tf.reduce_mean(entropy) self.global_step = tf.Variable(0, trainable=False, name='global_step', dtype=tf.int32) self.learning_rate = tf.train.polynomial_decay(lr, self.global_step, max_step, 1e-10, power=1.0) optimizer = tf.train.AdamOptimizer(learning_rate=self.learning_rate) self.update_batch = optimizer.minimize(self.loss) self.increment_step = tf.assign(self.global_step, self.global_step + 1) class ContinuousControlModel(PPOModel): def __init__(self, lr, s_size, a_size, h_size, epsilon, beta, max_step): """ Creates Continuous Control Actor-Critic model. :param s_size: State-space size :param a_size: Action-space size :param h_size: Hidden layer size """ self.state_in = tf.placeholder(shape=[None, s_size], dtype=tf.float32, name='state') self.batch_size = tf.placeholder(shape=None, dtype=tf.int32, name='batch_size') hidden_policy = tf.layers.dense(self.state_in, h_size, use_bias=False, activation=tf.nn.tanh) hidden_value = tf.layers.dense(self.state_in, h_size, use_bias=False, activation=tf.nn.tanh) hidden_policy_2 = tf.layers.dense(hidden_policy, h_size, use_bias=False, activation=tf.nn.tanh) hidden_value_2 = tf.layers.dense(hidden_value, h_size, use_bias=False, activation=tf.nn.tanh) self.mu = tf.layers.dense(hidden_policy_2, a_size, activation=None, use_bias=False, kernel_initializer=c_layers.variance_scaling_initializer(factor=0.1)) self.log_sigma_sq = tf.Variable(tf.zeros([a_size])) self.sigma_sq = tf.exp(self.log_sigma_sq) self.epsilon = tf.placeholder(shape=[None, a_size], dtype=tf.float32, name='epsilon') self.output = self.mu + tf.sqrt(self.sigma_sq) * self.epsilon self.output = tf.identity(self.output, name='action') a = tf.exp(-1 * tf.pow(tf.stop_gradient(self.output) - self.mu, 2) / (2 * self.sigma_sq)) b = 1 / tf.sqrt(2 * self.sigma_sq * np.pi) self.probs = a * b self.entropy = tf.reduce_sum(0.5 * tf.log(2 * np.pi * np.e * self.sigma_sq)) self.value = tf.layers.dense(hidden_value_2, 1, activation=None, use_bias=False) self.old_probs = tf.placeholder(shape=[None, a_size], dtype=tf.float32, name='old_probabilities') PPOModel.__init__(self, self.probs, self.old_probs, self.value, self.entropy, 0.0, epsilon, lr, max_step) class DiscreteControlModel(PPOModel): def __init__(self, lr, s_size, a_size, h_size, epsilon, beta, max_step): """ Creates Discrete Control Actor-Critic model. :param s_size: State-space size :param a_size: Action-space size :param h_size: Hidden layer size """ self.state_in = tf.placeholder(shape=[None, s_size], dtype=tf.float32, name='state') self.batch_size = tf.placeholder(shape=None, dtype=tf.int32, name='batch_size') hidden_1 = tf.layers.dense(self.state_in, h_size, use_bias=False, activation=tf.nn.elu) hidden_2 = tf.layers.dense(hidden_1, h_size, use_bias=False, activation=tf.nn.elu) self.policy = tf.layers.dense(hidden_2, a_size, activation=None, use_bias=False, kernel_initializer=c_layers.variance_scaling_initializer(factor=0.1)) self.probs = tf.nn.softmax(self.policy) self.action = tf.multinomial(self.policy, 1) self.output = tf.identity(self.action, name='action') self.value = tf.layers.dense(hidden_2, 1, activation=None, use_bias=False) self.entropy = -tf.reduce_sum(self.probs * tf.log(self.probs + 1e-10), axis=1) self.action_holder = tf.placeholder(shape=[None], dtype=tf.int32) self.selected_actions = c_layers.one_hot_encoding(self.action_holder, a_size) self.old_probs = tf.placeholder(shape=[None, a_size], dtype=tf.float32, name='old_probabilities') self.responsible_probs = tf.reduce_sum(self.probs * self.selected_actions, axis=1) self.old_responsible_probs = tf.reduce_sum(self.old_probs * self.selected_actions, axis=1) PPOModel.__init__(self, self.responsible_probs, self.old_responsible_probs, self.value, self.entropy, beta, epsilon, lr, max_step) class VisualDiscreteControlModel(PPOModel): def __init__(self, lr, o_size_h, o_size_w, a_size, h_size, epsilon, beta, max_step): """ Creates Discrete Control Actor-Critic model for use with visual observations (images). :param o_size_h: Observation height. :param o_size_w: Observation width. :param a_size: Action-space size. :param h_size: Hidden layer size. """ self.observation_in = tf.placeholder(shape=[None, o_size_h, o_size_w, 1], dtype=tf.float32, name='observation_0') self.conv1 = tf.layers.conv2d(self.observation_in, 32, kernel_size=[3, 3], strides=[2, 2], use_bias=False, activation=tf.nn.elu) self.conv2 = tf.layers.conv2d(self.conv1, 64, kernel_size=[3, 3], strides=[2, 2], use_bias=False, activation=tf.nn.elu) self.batch_size = tf.placeholder(shape=None, dtype=tf.int32) hidden = tf.layers.dense(c_layers.flatten(self.conv2), h_size, use_bias=False, activation=tf.nn.elu) self.policy = tf.layers.dense(hidden, a_size, activation=None, use_bias=False, kernel_initializer=c_layers.variance_scaling_initializer(factor=0.1)) self.probs = tf.nn.softmax(self.policy) self.action = tf.multinomial(self.policy, 1) self.output = tf.identity(self.action, name='action') self.value = tf.layers.dense(hidden, 1, activation=None, use_bias=False) self.entropy = -tf.reduce_sum(self.probs * tf.log(self.probs + 1e-10), axis=1) self.action_holder = tf.placeholder(shape=[None], dtype=tf.int32) self.selected_actions = c_layers.one_hot_encoding(self.action_holder, a_size) self.old_probs = tf.placeholder(shape=[None, a_size], dtype=tf.float32, name='old_probabilities') self.responsible_probs = tf.reduce_sum(self.probs * self.selected_actions, axis=1) self.old_responsible_probs = tf.reduce_sum(self.old_probs * self.selected_actions, axis=1) PPOModel.__init__(self, self.responsible_probs, self.old_responsible_probs, self.value, self.entropy, beta, epsilon, lr, max_step)