tensorforce.models package¶
Submodules¶
tensorforce.models.constant_model module¶
-
class
tensorforce.models.constant_model.
ConstantModel
(states_spec, actions_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, action_values)¶ Bases:
tensorforce.models.model.Model
Utility class to return constant actions of a desired shape and with given bounds.
-
tf_actions_and_internals
(states, internals, update, deterministic)¶
-
tf_loss_per_instance
(states, internals, actions, terminal, reward, update)¶
-
tensorforce.models.distribution_model module¶
-
class
tensorforce.models.distribution_model.
DistributionModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization)¶ Bases:
tensorforce.models.model.Model
Base class for models using distributions parameterized by a neural network.
-
create_distributions
()¶
-
static
get_distributions_summaries
(distributions)¶
-
static
get_distributions_variables
(distributions, include_non_trainable=False)¶
-
get_optimizer_kwargs
(states, internals, actions, terminal, reward, update)¶
-
get_summaries
()¶
-
get_variables
(include_non_trainable=False)¶
-
initialize
(custom_getter)¶
-
tf_actions_and_internals
(states, internals, update, deterministic)¶
-
tf_kl_divergence
(states, internals, update)¶
-
tf_regularization_losses
(states, internals, update)¶
-
tensorforce.models.model module¶
The Model
class coordinates the creation and execution of all TensorFlow operations within a model.
It implements the reset
, act
and update
functions, which give the interface the Agent
class
communicates with, and which should not need to be overwritten. Instead, the following TensorFlow
functions need to be implemented:
tf_actions_and_internals(states, internals, deterministic)
returning the batch of- actions and successor internal states.
tf_loss_per_instance(states, internals, actions, terminal, reward)
returning the loss- per instance for a batch.
Further, the following TensorFlow functions should be extended accordingly:
initialize(custom_getter)
defining TensorFlow placeholders/functions and adding internal states.get_variables()
returning the list of TensorFlow variables (to be optimized) of this model.tf_regularization_losses(states, internals)
returning a dict of regularization losses.get_optimizer_kwargs(states, internals, actions, terminal, reward)
returning a dict of potential- arguments (argument-free functions) to the optimizer.
Finally, the following TensorFlow functions can be useful in some cases:
preprocess_states(states)
for state preprocessing, returning the processed batch of states.action_exploration(action, exploration, action_spec)
for action postprocessing (e.g. exploration), returning the processed batch of actions.preprocess_reward(states, internals, terminal, reward)
for reward preprocessing (e.g. reward normalization), returning the processed batch of rewards.create_output_operations(states, internals, actions, terminal, reward, deterministic)
for further output operations, similar to the two above forModel.act
andModel.update
.tf_optimization(states, internals, actions, terminal, reward)
for further optimization operations (e.g. the baseline update in aPGModel
or the target network update in aQModel
), returning a single grouped optimization operation.
-
class
tensorforce.models.model.
Model
(states_spec, actions_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec)¶ Bases:
object
Base class for all (TensorFlow-based) models.
-
act
(states, internals, deterministic=False)¶
-
close
()¶
-
create_output_operations
(states, internals, actions, terminal, reward, update, deterministic)¶ Calls all the relevant TensorFlow functions for this model and hence creates all the TensorFlow operations involved.
Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- actions – Dict of action tensors.
- terminal – Terminal boolean tensor.
- reward – Reward tensor.
- update – Boolean tensor indicating whether this call happens during an update.
- deterministic – Boolean tensor indicating whether action should be chosen deterministically.
-
get_optimizer_kwargs
(states, internals, actions, terminal, reward, update)¶ Returns the optimizer arguments including the time, the list of variables to optimize, and various argument-free functions (in particular
fn_loss
returning the combined 0-dim batch loss tensor) which the optimizer might require to perform an update step.Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- actions – Dict of action tensors.
- terminal – Terminal boolean tensor.
- reward – Reward tensor.
- update – Boolean tensor indicating whether this call happens during an update.
Returns: Loss tensor of the size of the batch.
-
get_summaries
()¶ Returns the TensorFlow summaries reported by the model
Returns: List of summaries
-
get_variables
(include_non_trainable=False)¶ Returns the TensorFlow variables used by the model.
Returns: List of variables.
-
initialize
(custom_getter)¶ Creates the TensorFlow placeholders and functions for this model. Moreover adds the internal state placeholders and initialization values to the model.
Parameters: custom_getter – The custom_getter_
object to use fortf.make_template
when creating TensorFlow functions.
-
observe
(terminal, reward)¶
-
reset
()¶ Resets the model to its initial state on episode start.
Returns: Current episode and timestep counter, and a list containing the internal states initializations.
-
restore
(directory=None, file=None)¶ Restore TensorFlow model. If no checkpoint file is given, the latest checkpoint is restored. If no checkpoint directory is given, the model’s default saver directory is used (unless file specifies the entire path).
Parameters: - directory – Optional checkpoint directory.
- file – Optional checkpoint file, or path if directory not given.
-
save
(directory=None, append_timestep=True)¶ Save TensorFlow model. If no checkpoint directory is given, the model’s default saver directory is used. Optionally appends current timestep to prevent overwriting previous checkpoint files. Turn off to be able to load model from the same given path argument as given here.
Parameters: - directory – Optional checkpoint directory.
- append_timestep – Appends the current timestep to the checkpoint file if true.
Returns: Checkpoint path were the model was saved.
-
setup
()¶ Sets up the TensorFlow model graph and initializes the TensorFlow session.
-
tf_action_exploration
(action, exploration, action_spec)¶ Applies optional exploration to the action.
-
tf_actions_and_internals
(states, internals, update, deterministic)¶ Creates the TensorFlow operations for retrieving the actions (and posterior internal states) in reaction to the given input states (and prior internal states).
Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- update – Boolean tensor indicating whether this call happens during an update.
- deterministic – Boolean tensor indicating whether action should be chosen deterministically.
Returns: Actions and list of posterior internal state tensors.
-
tf_discounted_cumulative_reward
(terminal, reward, discount, final_reward=0.0)¶ Creates the TensorFlow operations for calculating the discounted cumulative rewards for a given sequence of rewards.
Parameters: - terminal – Terminal boolean tensor.
- reward – Reward tensor.
- discount – Discount factor.
- final_reward – Last reward value in the sequence.
Returns: Discounted cumulative reward tensor.
-
tf_loss
(states, internals, actions, terminal, reward, update)¶
-
tf_loss_per_instance
(states, internals, actions, terminal, reward, update)¶ Creates the TensorFlow operations for calculating the loss per batch instance of the given input states and actions.
Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- actions – Dict of action tensors.
- terminal – Terminal boolean tensor.
- reward – Reward tensor.
- update – Boolean tensor indicating whether this call happens during an update.
Returns: Loss tensor.
-
tf_optimization
(states, internals, actions, terminal, reward, update)¶ Creates the TensorFlow operations for performing an optimization update step based on the given input states and actions batch.
Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- actions – Dict of action tensors.
- terminal – Terminal boolean tensor.
- reward – Reward tensor.
- update – Boolean tensor indicating whether this call happens during an update.
Returns: The optimization operation.
-
tf_preprocess_reward
(states, internals, terminal, reward)¶ Applies optional pre-processing to the reward.
-
tf_preprocess_states
(states)¶ Applies optional pre-processing to the states.
-
tf_regularization_losses
(states, internals, update)¶ Creates the TensorFlow operations for calculating the regularization losses for the given input states.
Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- update – Boolean tensor indicating whether this call happens during an update.
Returns: Dict of regularization loss tensors.
-
update
(states, internals, actions, terminal, reward, return_loss_per_instance=False)¶
-
tensorforce.models.pg_log_prob_model module¶
-
class
tensorforce.models.pg_log_prob_model.
PGLogProbModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization, baseline_mode, baseline, baseline_optimizer, gae_lambda)¶ Bases:
tensorforce.models.pg_model.PGModel
Policy gradient model based on computing log likelihoods, e.g. VPG.
-
tf_pg_loss_per_instance
(states, internals, actions, terminal, reward, update)¶
-
tensorforce.models.pg_model module¶
-
class
tensorforce.models.pg_model.
PGModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization, baseline_mode, baseline, baseline_optimizer, gae_lambda)¶ Bases:
tensorforce.models.distribution_model.DistributionModel
Base class for policy gradient models. It optionally defines a baseline and handles its optimization. It implements the
tf_loss_per_instance
function, but requires subclasses to implementtf_pg_loss_per_instance
.-
get_summaries
()¶
-
get_variables
(include_non_trainable=False)¶
-
initialize
(custom_getter)¶
-
tf_loss_per_instance
(states, internals, actions, terminal, reward, update)¶
-
tf_optimization
(states, internals, actions, terminal, reward, update)¶
-
tf_pg_loss_per_instance
(states, internals, actions, terminal, reward, update)¶ Creates the TensorFlow operations for calculating the (policy-gradient-specific) loss per batch instance of the given input states and actions, after the specified reward/advantage calculations.
Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- actions – Dict of action tensors.
- terminal – Terminal boolean tensor.
- reward – Reward tensor.
- update – Boolean tensor indicating whether this call happens during an update.
Returns: Loss tensor.
-
tf_regularization_losses
(states, internals, update)¶
-
tf_reward_estimation
(states, internals, terminal, reward, update)¶
-
tensorforce.models.pg_prob_ratio_model module¶
-
class
tensorforce.models.pg_prob_ratio_model.
PGProbRatioModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization, baseline_mode, baseline, baseline_optimizer, gae_lambda, likelihood_ratio_clipping)¶ Bases:
tensorforce.models.pg_model.PGModel
Policy gradient model based on computing likelihood ratios, e.g. TRPO and PPO.
-
get_optimizer_kwargs
(states, actions, terminal, reward, internals, update)¶
-
initialize
(custom_getter)¶
-
tf_compare
(states, internals, actions, terminal, reward, update, reference)¶
-
tf_pg_loss_per_instance
(states, internals, actions, terminal, reward, update)¶
-
tf_reference
(states, internals, actions, update)¶
-
tensorforce.models.q_demo_model module¶
-
class
tensorforce.models.q_demo_model.
QDemoModel
(states_spec, actions_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, network_spec, distributions_spec, entropy_regularization, target_sync_frequency, target_update_weight, double_q_model, huber_loss, random_sampling_fix, expert_margin, supervised_weight)¶ Bases:
tensorforce.models.q_model.QModel
Model for deep Q-learning from demonstration. Principal structure similar to double deep Q-networks but uses additional loss terms for demo data.
-
create_output_operations
(states, internals, actions, terminal, reward, update, deterministic)¶
-
demonstration_update
(states, internals, actions, terminal, reward)¶
-
initialize
(custom_getter)¶
-
tf_demo_loss
(states, actions, terminal, reward, internals, update)¶
-
tf_demo_optimization
(states, internals, actions, terminal, reward, update)¶
-
tensorforce.models.q_model module¶
-
class
tensorforce.models.q_model.
QModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization, target_sync_frequency, target_update_weight, double_q_model, huber_loss, random_sampling_fix)¶ Bases:
tensorforce.models.distribution_model.DistributionModel
Q-value model.
-
get_summaries
()¶
-
get_variables
(include_non_trainable=False)¶
-
initialize
(custom_getter)¶
-
tf_loss_per_instance
(states, internals, actions, terminal, reward, update)¶
-
tf_optimization
(states, internals, actions, terminal, reward, update)¶
-
tf_q_delta
(q_value, next_q_value, terminal, reward)¶ Creates the deltas (or advantage) of the Q values.
Returns: A list of deltas per action
-
tf_q_value
(embedding, distr_params, action, name)¶
-
update
(states, internals, actions, terminal, reward, return_loss_per_instance=False)¶
-
tensorforce.models.q_naf_model module¶
-
class
tensorforce.models.q_naf_model.
QNAFModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization, target_sync_frequency, target_update_weight, double_q_model, huber_loss, random_sampling_fix)¶ Bases:
tensorforce.models.q_model.QModel
-
get_variables
(include_non_trainable=False)¶
-
initialize
(custom_getter)¶
-
tf_loss_per_instance
(states, internals, actions, terminal, reward, update)¶
-
tf_q_value
(embedding, distr_params, action, name)¶
-
tf_regularization_losses
(states, internals, update)¶
-
tensorforce.models.q_nstep_model module¶
-
class
tensorforce.models.q_nstep_model.
QNstepModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization, target_sync_frequency, target_update_weight, double_q_model, huber_loss, random_sampling_fix)¶ Bases:
tensorforce.models.q_model.QModel
Deep Q network using n-step rewards as described in Asynchronous Methods for Reinforcement Learning.
-
tf_q_delta
(q_value, next_q_value, terminal, reward)¶
-
tensorforce.models.random_model module¶
-
class
tensorforce.models.random_model.
RandomModel
(states_spec, actions_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec)¶ Bases:
tensorforce.models.model.Model
Utility class to return random actions of a desired shape and with given bounds.
-
tf_actions_and_internals
(states, internals, update, deterministic)¶
-
tf_loss_per_instance
(states, internals, actions, terminal, reward, update)¶
-
Module contents¶
-
class
tensorforce.models.
Model
(states_spec, actions_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec)¶ Bases:
object
Base class for all (TensorFlow-based) models.
-
act
(states, internals, deterministic=False)¶
-
close
()¶
-
create_output_operations
(states, internals, actions, terminal, reward, update, deterministic)¶ Calls all the relevant TensorFlow functions for this model and hence creates all the TensorFlow operations involved.
Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- actions – Dict of action tensors.
- terminal – Terminal boolean tensor.
- reward – Reward tensor.
- update – Boolean tensor indicating whether this call happens during an update.
- deterministic – Boolean tensor indicating whether action should be chosen deterministically.
-
get_optimizer_kwargs
(states, internals, actions, terminal, reward, update)¶ Returns the optimizer arguments including the time, the list of variables to optimize, and various argument-free functions (in particular
fn_loss
returning the combined 0-dim batch loss tensor) which the optimizer might require to perform an update step.Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- actions – Dict of action tensors.
- terminal – Terminal boolean tensor.
- reward – Reward tensor.
- update – Boolean tensor indicating whether this call happens during an update.
Returns: Loss tensor of the size of the batch.
-
get_summaries
()¶ Returns the TensorFlow summaries reported by the model
Returns: List of summaries
-
get_variables
(include_non_trainable=False)¶ Returns the TensorFlow variables used by the model.
Returns: List of variables.
-
initialize
(custom_getter)¶ Creates the TensorFlow placeholders and functions for this model. Moreover adds the internal state placeholders and initialization values to the model.
Parameters: custom_getter – The custom_getter_
object to use fortf.make_template
when creating TensorFlow functions.
-
observe
(terminal, reward)¶
-
reset
()¶ Resets the model to its initial state on episode start.
Returns: Current episode and timestep counter, and a list containing the internal states initializations.
-
restore
(directory=None, file=None)¶ Restore TensorFlow model. If no checkpoint file is given, the latest checkpoint is restored. If no checkpoint directory is given, the model’s default saver directory is used (unless file specifies the entire path).
Parameters: - directory – Optional checkpoint directory.
- file – Optional checkpoint file, or path if directory not given.
-
save
(directory=None, append_timestep=True)¶ Save TensorFlow model. If no checkpoint directory is given, the model’s default saver directory is used. Optionally appends current timestep to prevent overwriting previous checkpoint files. Turn off to be able to load model from the same given path argument as given here.
Parameters: - directory – Optional checkpoint directory.
- append_timestep – Appends the current timestep to the checkpoint file if true.
Returns: Checkpoint path were the model was saved.
-
setup
()¶ Sets up the TensorFlow model graph and initializes the TensorFlow session.
-
tf_action_exploration
(action, exploration, action_spec)¶ Applies optional exploration to the action.
-
tf_actions_and_internals
(states, internals, update, deterministic)¶ Creates the TensorFlow operations for retrieving the actions (and posterior internal states) in reaction to the given input states (and prior internal states).
Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- update – Boolean tensor indicating whether this call happens during an update.
- deterministic – Boolean tensor indicating whether action should be chosen deterministically.
Returns: Actions and list of posterior internal state tensors.
-
tf_discounted_cumulative_reward
(terminal, reward, discount, final_reward=0.0)¶ Creates the TensorFlow operations for calculating the discounted cumulative rewards for a given sequence of rewards.
Parameters: - terminal – Terminal boolean tensor.
- reward – Reward tensor.
- discount – Discount factor.
- final_reward – Last reward value in the sequence.
Returns: Discounted cumulative reward tensor.
-
tf_loss
(states, internals, actions, terminal, reward, update)¶
-
tf_loss_per_instance
(states, internals, actions, terminal, reward, update)¶ Creates the TensorFlow operations for calculating the loss per batch instance of the given input states and actions.
Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- actions – Dict of action tensors.
- terminal – Terminal boolean tensor.
- reward – Reward tensor.
- update – Boolean tensor indicating whether this call happens during an update.
Returns: Loss tensor.
-
tf_optimization
(states, internals, actions, terminal, reward, update)¶ Creates the TensorFlow operations for performing an optimization update step based on the given input states and actions batch.
Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- actions – Dict of action tensors.
- terminal – Terminal boolean tensor.
- reward – Reward tensor.
- update – Boolean tensor indicating whether this call happens during an update.
Returns: The optimization operation.
-
tf_preprocess_reward
(states, internals, terminal, reward)¶ Applies optional pre-processing to the reward.
-
tf_preprocess_states
(states)¶ Applies optional pre-processing to the states.
-
tf_regularization_losses
(states, internals, update)¶ Creates the TensorFlow operations for calculating the regularization losses for the given input states.
Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- update – Boolean tensor indicating whether this call happens during an update.
Returns: Dict of regularization loss tensors.
-
update
(states, internals, actions, terminal, reward, return_loss_per_instance=False)¶
-
-
class
tensorforce.models.
DistributionModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization)¶ Bases:
tensorforce.models.model.Model
Base class for models using distributions parameterized by a neural network.
-
create_distributions
()¶
-
static
get_distributions_summaries
(distributions)¶
-
static
get_distributions_variables
(distributions, include_non_trainable=False)¶
-
get_optimizer_kwargs
(states, internals, actions, terminal, reward, update)¶
-
get_summaries
()¶
-
get_variables
(include_non_trainable=False)¶
-
initialize
(custom_getter)¶
-
tf_actions_and_internals
(states, internals, update, deterministic)¶
-
tf_kl_divergence
(states, internals, update)¶
-
tf_regularization_losses
(states, internals, update)¶
-
-
class
tensorforce.models.
PGModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization, baseline_mode, baseline, baseline_optimizer, gae_lambda)¶ Bases:
tensorforce.models.distribution_model.DistributionModel
Base class for policy gradient models. It optionally defines a baseline and handles its optimization. It implements the
tf_loss_per_instance
function, but requires subclasses to implementtf_pg_loss_per_instance
.-
get_summaries
()¶
-
get_variables
(include_non_trainable=False)¶
-
initialize
(custom_getter)¶
-
tf_loss_per_instance
(states, internals, actions, terminal, reward, update)¶
-
tf_optimization
(states, internals, actions, terminal, reward, update)¶
-
tf_pg_loss_per_instance
(states, internals, actions, terminal, reward, update)¶ Creates the TensorFlow operations for calculating the (policy-gradient-specific) loss per batch instance of the given input states and actions, after the specified reward/advantage calculations.
Parameters: - states – Dict of state tensors.
- internals – List of prior internal state tensors.
- actions – Dict of action tensors.
- terminal – Terminal boolean tensor.
- reward – Reward tensor.
- update – Boolean tensor indicating whether this call happens during an update.
Returns: Loss tensor.
-
tf_regularization_losses
(states, internals, update)¶
-
tf_reward_estimation
(states, internals, terminal, reward, update)¶
-
-
class
tensorforce.models.
PGProbRatioModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization, baseline_mode, baseline, baseline_optimizer, gae_lambda, likelihood_ratio_clipping)¶ Bases:
tensorforce.models.pg_model.PGModel
Policy gradient model based on computing likelihood ratios, e.g. TRPO and PPO.
-
get_optimizer_kwargs
(states, actions, terminal, reward, internals, update)¶
-
initialize
(custom_getter)¶
-
tf_compare
(states, internals, actions, terminal, reward, update, reference)¶
-
tf_pg_loss_per_instance
(states, internals, actions, terminal, reward, update)¶
-
tf_reference
(states, internals, actions, update)¶
-
-
class
tensorforce.models.
PGLogProbModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization, baseline_mode, baseline, baseline_optimizer, gae_lambda)¶ Bases:
tensorforce.models.pg_model.PGModel
Policy gradient model based on computing log likelihoods, e.g. VPG.
-
tf_pg_loss_per_instance
(states, internals, actions, terminal, reward, update)¶
-
-
class
tensorforce.models.
QModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization, target_sync_frequency, target_update_weight, double_q_model, huber_loss, random_sampling_fix)¶ Bases:
tensorforce.models.distribution_model.DistributionModel
Q-value model.
-
get_summaries
()¶
-
get_variables
(include_non_trainable=False)¶
-
initialize
(custom_getter)¶
-
tf_loss_per_instance
(states, internals, actions, terminal, reward, update)¶
-
tf_optimization
(states, internals, actions, terminal, reward, update)¶
-
tf_q_delta
(q_value, next_q_value, terminal, reward)¶ Creates the deltas (or advantage) of the Q values.
Returns: A list of deltas per action
-
tf_q_value
(embedding, distr_params, action, name)¶
-
update
(states, internals, actions, terminal, reward, return_loss_per_instance=False)¶
-
-
class
tensorforce.models.
QNstepModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization, target_sync_frequency, target_update_weight, double_q_model, huber_loss, random_sampling_fix)¶ Bases:
tensorforce.models.q_model.QModel
Deep Q network using n-step rewards as described in Asynchronous Methods for Reinforcement Learning.
-
tf_q_delta
(q_value, next_q_value, terminal, reward)¶
-
-
class
tensorforce.models.
QNAFModel
(states_spec, actions_spec, network_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, distributions_spec, entropy_regularization, target_sync_frequency, target_update_weight, double_q_model, huber_loss, random_sampling_fix)¶ Bases:
tensorforce.models.q_model.QModel
-
get_variables
(include_non_trainable=False)¶
-
initialize
(custom_getter)¶
-
tf_loss_per_instance
(states, internals, actions, terminal, reward, update)¶
-
tf_q_value
(embedding, distr_params, action, name)¶
-
tf_regularization_losses
(states, internals, update)¶
-
-
class
tensorforce.models.
QDemoModel
(states_spec, actions_spec, device, session_config, scope, saver_spec, summary_spec, distributed_spec, optimizer, discount, variable_noise, states_preprocessing_spec, explorations_spec, reward_preprocessing_spec, network_spec, distributions_spec, entropy_regularization, target_sync_frequency, target_update_weight, double_q_model, huber_loss, random_sampling_fix, expert_margin, supervised_weight)¶ Bases:
tensorforce.models.q_model.QModel
Model for deep Q-learning from demonstration. Principal structure similar to double deep Q-networks but uses additional loss terms for demo data.
-
create_output_operations
(states, internals, actions, terminal, reward, update, deterministic)¶
-
demonstration_update
(states, internals, actions, terminal, reward)¶
-
initialize
(custom_getter)¶
-
tf_demo_loss
(states, actions, terminal, reward, internals, update)¶
-
tf_demo_optimization
(states, internals, actions, terminal, reward, update)¶
-