Unity 机器学习代理工具包 (ML-Agents) 是一个开源项目,它使游戏和模拟能够作为训练智能代理的环境。
您最多选择25个主题 主题必须以中文或者字母或数字开头,可以包含连字符 (-),并且长度不得超过35个字符
 
 
 
 
 

1362 行
60 KiB

using System;
using System.Collections.Generic;
using System.Collections.ObjectModel;
using UnityEngine;
using Unity.Barracuda;
using Unity.MLAgents.Actuators;
using Unity.MLAgents.Sensors;
using Unity.MLAgents.Sensors.Reflection;
using Unity.MLAgents.Demonstrations;
using Unity.MLAgents.Policies;
using UnityEngine.Serialization;
namespace Unity.MLAgents
{
/// <summary>
/// Struct that contains all the information for an Agent, including its
/// observations, actions and current status.
/// </summary>
internal struct AgentInfo
{
/// <summary>
/// Keeps track of the last actions taken by the Brain.
/// </summary>
public ActionBuffers storedActions;
/// <summary>
/// For discrete control, specifies the actions that the agent cannot take.
/// An element of the mask array is <c>true</c> if the action is prohibited.
/// </summary>
public bool[] discreteActionMasks;
/// <summary>
/// The current agent reward.
/// </summary>
public float reward;
/// <summary>
/// Whether the agent is done or not.
/// </summary>
public bool done;
/// <summary>
/// Whether the agent has reached its max step count for this episode.
/// </summary>
public bool maxStepReached;
/// <summary>
/// Episode identifier each agent receives at every reset. It is used
/// to separate between different agents in the environment.
/// </summary>
public int episodeId;
public void ClearActions()
{
storedActions.Clear();
}
public void CopyActions(ActionBuffers actionBuffers)
{
var continuousActions = storedActions.ContinuousActions;
for (var i = 0; i < actionBuffers.ContinuousActions.Length; i++)
{
continuousActions[i] = actionBuffers.ContinuousActions[i];
}
var discreteActions = storedActions.DiscreteActions;
for (var i = 0; i < actionBuffers.DiscreteActions.Length; i++)
{
discreteActions[i] = actionBuffers.DiscreteActions[i];
}
}
}
/// <summary>
/// An agent is an actor that can observe its environment, decide on the
/// best course of action using those observations, and execute those actions
/// within the environment.
/// </summary>
/// <remarks>
/// Use the Agent class as the subclass for implementing your own agents. Add
/// your Agent implementation to a [GameObject] in the [Unity scene] that serves
/// as the agent's environment.
///
/// Agents in an environment operate in *steps*. At each step, an agent collects observations,
/// passes them to its decision-making policy, and receives an action vector in response.
///
/// Agents make observations using <see cref="ISensor"/> implementations. The ML-Agents
/// API provides implementations for visual observations (<see cref="CameraSensor"/>)
/// raycast observations (<see cref="RayPerceptionSensor"/>), and arbitrary
/// data observations (<see cref="VectorSensor"/>). You can add the
/// <see cref="CameraSensorComponent"/> and <see cref="RayPerceptionSensorComponent2D"/> or
/// <see cref="RayPerceptionSensorComponent3D"/> components to an agent's [GameObject] to use
/// those sensor types. You can implement the <see cref="CollectObservations(VectorSensor)"/>
/// function in your Agent subclass to use a vector observation. The Agent class calls this
/// function before it uses the observation vector to make a decision. (If you only use
/// visual or raycast observations, you do not need to implement
/// <see cref="CollectObservations"/>.)
///
/// Assign a decision making policy to an agent using a <see cref="BehaviorParameters"/>
/// component attached to the agent's [GameObject]. The <see cref="BehaviorType"/> setting
/// determines how decisions are made:
///
/// * <see cref="BehaviorType.Default"/>: decisions are made by the external process,
/// when connected. Otherwise, decisions are made using inference. If no inference model
/// is specified in the BehaviorParameters component, then heuristic decision
/// making is used.
/// * <see cref="BehaviorType.InferenceOnly"/>: decisions are always made using the trained
/// model specified in the <see cref="BehaviorParameters"/> component.
/// * <see cref="BehaviorType.HeuristicOnly"/>: when a decision is needed, the agent's
/// <see cref="Heuristic(in ActionBuffers)"/> function is called. Your implementation is responsible for
/// providing the appropriate action.
///
/// To trigger an agent decision automatically, you can attach a <see cref="DecisionRequester"/>
/// component to the Agent game object. You can also call the agent's <see cref="RequestDecision"/>
/// function manually. You only need to call <see cref="RequestDecision"/> when the agent is
/// in a position to act upon the decision. In many cases, this will be every [FixedUpdate]
/// callback, but could be less frequent. For example, an agent that hops around its environment
/// can only take an action when it touches the ground, so several frames might elapse between
/// one decision and the need for the next.
///
/// Use the <see cref="OnActionReceived(ActionBuffers)"/> function to implement the actions your agent can take,
/// such as moving to reach a goal or interacting with its environment.
///
/// When you call <see cref="EndEpisode"/> on an agent or the agent reaches its <see cref="MaxStep"/> count,
/// its current episode ends. You can reset the agent -- or remove it from the
/// environment -- by implementing the <see cref="OnEpisodeBegin"/> function. An agent also
/// becomes done when the <see cref="Academy"/> resets the environment, which only happens when
/// the <see cref="Academy"/> receives a reset signal from an external process via the
/// <see cref="Academy.Communicator"/>.
///
/// The Agent class extends the Unity [MonoBehaviour] class. You can implement the
/// standard [MonoBehaviour] functions as needed for your agent. Since an agent's
/// observations and actions typically take place during the [FixedUpdate] phase, you should
/// only use the [MonoBehaviour.Update] function for cosmetic purposes. If you override the [MonoBehaviour]
/// methods, [OnEnable()] or [OnDisable()], always call the base Agent class implementations.
///
/// You can implement the <see cref="Heuristic(in ActionBuffers)"/> function to specify agent actions using
/// your own heuristic algorithm. Implementing a heuristic function can be useful
/// for debugging. For example, you can use keyboard input to select agent actions in
/// order to manually control an agent's behavior.
///
/// Note that you can change the inference model assigned to an agent at any step
/// by calling <see cref="SetModel"/>.
///
/// See [Agents] and [Reinforcement Learning in Unity] in the [Unity ML-Agents Toolkit manual] for
/// more information on creating and training agents.
///
/// For sample implementations of agent behavior, see the examples available in the
/// [Unity ML-Agents Toolkit] on Github.
///
/// [MonoBehaviour]: https://docs.unity3d.com/ScriptReference/MonoBehaviour.html
/// [GameObject]: https://docs.unity3d.com/Manual/GameObjects.html
/// [Unity scene]: https://docs.unity3d.com/Manual/CreatingScenes.html
/// [FixedUpdate]: https://docs.unity3d.com/ScriptReference/MonoBehaviour.FixedUpdate.html
/// [MonoBehaviour.Update]: https://docs.unity3d.com/ScriptReference/MonoBehaviour.Update.html
/// [OnEnable()]: https://docs.unity3d.com/ScriptReference/MonoBehaviour.OnEnable.html
/// [OnDisable()]: https://docs.unity3d.com/ScriptReference/MonoBehaviour.OnDisable.html]
/// [OnBeforeSerialize()]: https://docs.unity3d.com/ScriptReference/MonoBehaviour.OnBeforeSerialize.html
/// [OnAfterSerialize()]: https://docs.unity3d.com/ScriptReference/MonoBehaviour.OnAfterSerialize.html
/// [Agents]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/Learning-Environment-Design-Agents.md
/// [Reinforcement Learning in Unity]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/Learning-Environment-Design.md
/// [Unity ML-Agents Toolkit]: https://github.com/Unity-Technologies/ml-agents
/// [Unity ML-Agents Toolkit manual]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/Readme.md
///
/// </remarks>
[HelpURL("https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/" +
"docs/Learning-Environment-Design-Agents.md")]
[Serializable]
[RequireComponent(typeof(BehaviorParameters))]
public partial class Agent : MonoBehaviour, ISerializationCallbackReceiver, IActionReceiver, IHeuristicProvider
{
IPolicy m_Brain;
BehaviorParameters m_PolicyFactory;
/// This code is here to make the upgrade path for users using MaxStep
/// easier. We will hook into the Serialization code and make sure that
/// agentParameters.maxStep and this.maxStep are in sync.
[Serializable]
internal struct AgentParameters
{
public int maxStep;
}
[SerializeField]
[HideInInspector]
internal AgentParameters agentParameters;
[SerializeField]
[HideInInspector]
internal bool hasUpgradedFromAgentParameters;
/// <summary>
/// The maximum number of steps the agent takes before being done.
/// </summary>
/// <value>The maximum steps for an agent to take before it resets; or 0 for
/// unlimited steps.</value>
/// <remarks>
/// The max step value determines the maximum length of an agent's episodes.
/// Set to a positive integer to limit the episode length to that many steps.
/// Set to 0 for unlimited episode length.
///
/// When an episode ends and a new one begins, the Agent object's
/// <seealso cref="OnEpisodeBegin"/> function is called. You can implement
/// <see cref="OnEpisodeBegin"/> to reset the agent or remove it from the
/// environment. An agent's episode can also end if you call its <seealso cref="EndEpisode"/>
/// method or an external process resets the environment through the <see cref="Academy"/>.
///
/// Consider limiting the number of steps in an episode to avoid wasting time during
/// training. If you set the max step value to a reasonable estimate of the time it should
/// take to complete a task, then agents that haven’t succeeded in that time frame will
/// reset and start a new training episode rather than continue to fail.
/// </remarks>
/// <example>
/// To use a step limit when training while allowing agents to run without resetting
/// outside of training, you can set the max step to 0 in <see cref="Initialize"/>
/// if the <see cref="Academy"/> is not connected to an external process.
/// <code>
/// using Unity.MLAgents;
///
/// public class MyAgent : Agent
/// {
/// public override void Initialize()
/// {
/// if (!Academy.Instance.IsCommunicatorOn)
/// {
/// this.MaxStep = 0;
/// }
/// }
/// }
/// </code>
/// **Note:** in general, you should limit the differences between the code you execute
/// during training and the code you run during inference.
/// </example>
[FormerlySerializedAs("maxStep")]
[HideInInspector] public int MaxStep;
/// Current Agent information (message sent to Brain).
AgentInfo m_Info;
/// Represents the reward the agent accumulated during the current step.
/// It is reset to 0 at the beginning of every step.
/// Should be set to a positive value when the agent performs a "good"
/// action that we wish to reinforce/reward, and set to a negative value
/// when the agent performs a "bad" action that we wish to punish/deter.
/// Additionally, the magnitude of the reward should not exceed 1.0
float m_Reward;
/// Keeps track of the cumulative reward in this episode.
float m_CumulativeReward;
/// Whether or not the agent requests an action.
bool m_RequestAction;
/// Whether or not the agent requests a decision.
bool m_RequestDecision;
/// Keeps track of the number of steps taken by the agent in this episode.
/// Note that this value is different for each agent, and may not overlap
/// with the step counter in the Academy, since agents reset based on
/// their own experience.
int m_StepCount;
/// Number of times the Agent has completed an episode.
int m_CompletedEpisodes;
/// Episode identifier each agent receives. It is used
/// to separate between different agents in the environment.
/// This Id will be changed every time the Agent resets.
int m_EpisodeId;
/// Whether or not the Agent has been initialized already
bool m_Initialized;
/// Keeps track of the actions that are masked at each step.
DiscreteActionMasker m_ActionMasker;
/// <summary>
/// Set of DemonstrationWriters that the Agent will write its step information to.
/// If you use a DemonstrationRecorder component, this will automatically register its DemonstrationWriter.
/// You can also add your own DemonstrationWriter by calling
/// DemonstrationRecorder.AddDemonstrationWriterToAgent()
/// </summary>
internal ISet<DemonstrationWriter> DemonstrationWriters = new HashSet<DemonstrationWriter>();
/// <summary>
/// List of sensors used to generate observations.
/// Currently generated from attached SensorComponents, and a legacy VectorSensor
/// </summary>
internal List<ISensor> sensors;
/// <summary>
/// VectorSensor which is written to by AddVectorObs
/// </summary>
internal VectorSensor collectObservationsSensor;
private RecursionChecker m_CollectObservationsChecker = new RecursionChecker("CollectObservations");
private RecursionChecker m_OnEpisodeBeginChecker = new RecursionChecker("OnEpisodeBegin");
/// <summary>
/// List of IActuators that this Agent will delegate actions to if any exist.
/// </summary>
ActuatorManager m_ActuatorManager;
/// <summary>
/// VectorActuator which is used by default if no other sensors exist on this Agent. This VectorSensor will
/// delegate its actions to <see cref="OnActionReceived(ActionBuffers)"/> by default in order to keep backward compatibility
/// with the current behavior of Agent.
/// </summary>
IActuator m_VectorActuator;
/// <summary>
/// This is used to avoid allocation of a float array every frame if users are still using the old
/// OnActionReceived method.
/// </summary>
float[] m_LegacyActionCache;
/// <summary>
/// This is used to avoid allocation of a float array during legacy calls to Heuristic.
/// </summary>
float[] m_LegacyHeuristicCache;
/// <summary>
/// Called when the attached [GameObject] becomes enabled and active.
/// [GameObject]: https://docs.unity3d.com/Manual/GameObjects.html
/// </summary>
/// <remarks>
/// This function initializes the Agent instance, if it hasn't been initialized yet.
/// Always call the base Agent class version of this function if you implement `OnEnable()`
/// in your own Agent subclasses.
/// </remarks>
/// <example>
/// <code>
/// protected override void OnEnable()
/// {
/// base.OnEnable();
/// // additional OnEnable logic...
/// }
/// </code>
/// </example>
protected virtual void OnEnable()
{
LazyInitialize();
}
/// <summary>
/// Called by Unity immediately before serializing this object.
/// </summary>
/// <remarks>
/// The Agent class uses OnBeforeSerialize() for internal housekeeping. Call the
/// base class implementation if you need your own custom serialization logic.
///
/// See [OnBeforeSerialize] for more information.
///
/// [OnBeforeSerialize]: https://docs.unity3d.com/ScriptReference/ISerializationCallbackReceiver.OnAfterDeserialize.html
/// </remarks>
/// <example>
/// <code>
/// public new void OnBeforeSerialize()
/// {
/// base.OnBeforeSerialize();
/// // additional serialization logic...
/// }
/// </code>
/// </example>
public void OnBeforeSerialize()
{
// Manages a serialization upgrade issue from v0.13 to v0.14 where MaxStep moved
// from AgentParameters (since removed) to Agent
if (MaxStep == 0 && MaxStep != agentParameters.maxStep && !hasUpgradedFromAgentParameters)
{
MaxStep = agentParameters.maxStep;
}
hasUpgradedFromAgentParameters = true;
}
/// <summary>
/// Called by Unity immediately after deserializing this object.
/// </summary>
/// <remarks>
/// The Agent class uses OnAfterDeserialize() for internal housekeeping. Call the
/// base class implementation if you need your own custom deserialization logic.
///
/// See [OnAfterDeserialize] for more information.
///
/// [OnAfterDeserialize]: https://docs.unity3d.com/ScriptReference/ISerializationCallbackReceiver.OnAfterDeserialize.html
/// </remarks>
/// <example>
/// <code>
/// public new void OnAfterDeserialize()
/// {
/// base.OnAfterDeserialize();
/// // additional deserialization logic...
/// }
/// </code>
/// </example>
public void OnAfterDeserialize()
{
// Manages a serialization upgrade issue from v0.13 to v0.14 where MaxStep moved
// from AgentParameters (since removed) to Agent
if (MaxStep == 0 && MaxStep != agentParameters.maxStep && !hasUpgradedFromAgentParameters)
{
MaxStep = agentParameters.maxStep;
}
hasUpgradedFromAgentParameters = true;
}
/// <summary>
/// Initializes the agent. Can be safely called multiple times.
/// </summary>
/// <remarks>
/// This function calls your <seealso cref="Initialize"/> implementation, if one exists.
/// </remarks>
public void LazyInitialize()
{
if (m_Initialized)
{
return;
}
m_Initialized = true;
// Grab the "static" properties for the Agent.
m_EpisodeId = EpisodeIdCounter.GetEpisodeId();
m_PolicyFactory = GetComponent<BehaviorParameters>();
m_Info = new AgentInfo();
sensors = new List<ISensor>();
Academy.Instance.AgentIncrementStep += AgentIncrementStep;
Academy.Instance.AgentSendState += SendInfo;
Academy.Instance.DecideAction += DecideAction;
Academy.Instance.AgentAct += AgentStep;
Academy.Instance.AgentForceReset += _AgentReset;
using (TimerStack.Instance.Scoped("InitializeActuators"))
{
InitializeActuators();
}
m_Brain = m_PolicyFactory.GeneratePolicy(m_ActuatorManager.GetCombinedActionSpec(), m_ActuatorManager);
ResetData();
Initialize();
using (TimerStack.Instance.Scoped("InitializeSensors"))
{
InitializeSensors();
}
m_Info.storedActions = new ActionBuffers(
new float[m_ActuatorManager.NumContinuousActions],
new int[m_ActuatorManager.NumDiscreteActions]
);
// The first time the Academy resets, all Agents in the scene will be
// forced to reset through the <see cref="AgentForceReset"/> event.
// To avoid the Agent resetting twice, the Agents will not begin their
// episode when initializing until after the Academy had its first reset.
if (Academy.Instance.TotalStepCount != 0)
{
using (m_OnEpisodeBeginChecker.Start())
{
OnEpisodeBegin();
}
}
}
/// <summary>
/// The reason that the Agent has been set to "done".
/// </summary>
enum DoneReason
{
/// <summary>
/// The episode was ended manually by calling <see cref="EndEpisode"/>.
/// </summary>
DoneCalled,
/// <summary>
/// The max steps for the Agent were reached.
/// </summary>
MaxStepReached,
/// <summary>
/// The Agent was disabled.
/// </summary>
Disabled,
}
/// <summary>
/// Called when the attached [GameObject] becomes disabled and inactive.
/// [GameObject]: https://docs.unity3d.com/Manual/GameObjects.html
/// </summary>
/// <remarks>
/// Always call the base Agent class version of this function if you implement `OnDisable()`
/// in your own Agent subclasses.
/// </remarks>
/// <example>
/// <code>
/// protected override void OnDisable()
/// {
/// base.OnDisable();
/// // additional OnDisable logic...
/// }
/// </code>
/// </example>
/// <seealso cref="OnEnable"/>
protected virtual void OnDisable()
{
DemonstrationWriters.Clear();
// If Academy.Dispose has already been called, we don't need to unregister with it.
// We don't want to even try, because this will lazily create a new Academy!
if (Academy.IsInitialized)
{
Academy.Instance.AgentIncrementStep -= AgentIncrementStep;
Academy.Instance.AgentSendState -= SendInfo;
Academy.Instance.DecideAction -= DecideAction;
Academy.Instance.AgentAct -= AgentStep;
Academy.Instance.AgentForceReset -= _AgentReset;
NotifyAgentDone(DoneReason.Disabled);
}
m_Brain?.Dispose();
m_Initialized = false;
}
void NotifyAgentDone(DoneReason doneReason)
{
if (m_Info.done)
{
// The Agent was already marked as Done and should not be notified again
return;
}
m_Info.episodeId = m_EpisodeId;
m_Info.reward = m_Reward;
m_Info.done = true;
m_Info.maxStepReached = doneReason == DoneReason.MaxStepReached;
if (collectObservationsSensor != null)
{
// Make sure the latest observations are being passed to training.
collectObservationsSensor.Reset();
using (m_CollectObservationsChecker.Start())
{
CollectObservations(collectObservationsSensor);
}
}
// Request the last decision with no callbacks
// We request a decision so Python knows the Agent is done immediately
m_Brain?.RequestDecision(m_Info, sensors);
ResetSensors();
// We also have to write any to any DemonstationStores so that they get the "done" flag.
foreach (var demoWriter in DemonstrationWriters)
{
demoWriter.Record(m_Info, sensors);
}
if (doneReason != DoneReason.Disabled)
{
// We don't want to update the reward stats when the Agent is disabled, because this will make
// the rewards look lower than they actually are during shutdown.
m_CompletedEpisodes++;
UpdateRewardStats();
}
m_Reward = 0f;
m_CumulativeReward = 0f;
m_RequestAction = false;
m_RequestDecision = false;
m_Info.storedActions.Clear();
}
/// <summary>
/// Updates the Model assigned to this Agent instance.
/// </summary>
/// <remarks>
/// If the agent already has an assigned model, that model is replaced with the
/// the provided one. However, if you call this function with arguments that are
/// identical to the current parameters of the agent, then no changes are made.
///
/// **Note:** the <paramref name="behaviorName"/> parameter is ignored when not training.
/// The <paramref name="model"/> and <paramref name="inferenceDevice"/> parameters
/// are ignored when not using inference.
/// </remarks>
/// <param name="behaviorName"> The identifier of the behavior. This
/// will categorize the agent when training.
/// </param>
/// <param name="model"> The model to use for inference.</param>
/// <param name = "inferenceDevice"> Define the device on which the model
/// will be run.</param>
public void SetModel(
string behaviorName,
NNModel model,
InferenceDevice inferenceDevice = InferenceDevice.CPU)
{
if (behaviorName == m_PolicyFactory.BehaviorName &&
model == m_PolicyFactory.Model &&
inferenceDevice == m_PolicyFactory.InferenceDevice)
{
// If everything is the same, don't make any changes.
return;
}
NotifyAgentDone(DoneReason.Disabled);
m_PolicyFactory.Model = model;
m_PolicyFactory.InferenceDevice = inferenceDevice;
m_PolicyFactory.BehaviorName = behaviorName;
ReloadPolicy();
}
internal void ReloadPolicy()
{
if (!m_Initialized)
{
// If we haven't initialized yet, no need to make any changes now; they'll
// happen in LazyInitialize later.
return;
}
m_Brain?.Dispose();
m_Brain = m_PolicyFactory.GeneratePolicy(m_ActuatorManager.GetCombinedActionSpec(), m_ActuatorManager);
}
/// <summary>
/// Returns the current step counter (within the current episode).
/// </summary>
/// <returns>
/// Current step count.
/// </returns>
public int StepCount
{
get { return m_StepCount; }
}
/// <summary>
/// Returns the number of episodes that the Agent has completed (either <see cref="Agent.EndEpisode()"/>
/// was called, or maxSteps was reached).
/// </summary>
/// <returns>
/// Current episode count.
/// </returns>
public int CompletedEpisodes
{
get { return m_CompletedEpisodes; }
}
/// <summary>
/// Overrides the current step reward of the agent and updates the episode
/// reward accordingly.
/// </summary>
/// <remarks>
/// This function replaces any rewards given to the agent during the current step.
/// Use <see cref="AddReward(float)"/> to incrementally change the reward rather than
/// overriding it.
///
/// Typically, you assign rewards in the Agent subclass's <see cref="OnActionReceived(ActionBuffers)"/>
/// implementation after carrying out the received action and evaluating its success.
///
/// Rewards are used during reinforcement learning; they are ignored during inference.
///
/// See [Agents - Rewards] for general advice on implementing rewards and [Reward Signals]
/// for information about mixing reward signals from curiosity and Generative Adversarial
/// Imitation Learning (GAIL) with rewards supplied through this method.
///
/// [Agents - Rewards]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/Learning-Environment-Design-Agents.md#rewards
/// [Reward Signals]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/ML-Agents-Overview.md#a-quick-note-on-reward-signals
/// </remarks>
/// <param name="reward">The new value of the reward.</param>
public void SetReward(float reward)
{
#if DEBUG
Utilities.DebugCheckNanAndInfinity(reward, nameof(reward), nameof(SetReward));
#endif
m_CumulativeReward += (reward - m_Reward);
m_Reward = reward;
}
/// <summary>
/// Increments the step and episode rewards by the provided value.
/// </summary>
/// <remarks>Use a positive reward to reinforce desired behavior. You can use a
/// negative reward to penalize mistakes. Use <seealso cref="SetReward(float)"/> to
/// set the reward assigned to the current step with a specific value rather than
/// increasing or decreasing it.
///
/// Typically, you assign rewards in the Agent subclass's <see cref="IActionReceiver.OnActionReceived"/>
/// implementation after carrying out the received action and evaluating its success.
///
/// Rewards are used during reinforcement learning; they are ignored during inference.
///
/// See [Agents - Rewards] for general advice on implementing rewards and [Reward Signals]
/// for information about mixing reward signals from curiosity and Generative Adversarial
/// Imitation Learning (GAIL) with rewards supplied through this method.
///
/// [Agents - Rewards]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/Learning-Environment-Design-Agents.md#rewards
/// [Reward Signals]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/ML-Agents-Overview.md#a-quick-note-on-reward-signals
///</remarks>
/// <param name="increment">Incremental reward value.</param>
public void AddReward(float increment)
{
#if DEBUG
Utilities.DebugCheckNanAndInfinity(increment, nameof(increment), nameof(AddReward));
#endif
m_Reward += increment;
m_CumulativeReward += increment;
}
/// <summary>
/// Retrieves the episode reward for the Agent.
/// </summary>
/// <returns>The episode reward.</returns>
public float GetCumulativeReward()
{
return m_CumulativeReward;
}
void UpdateRewardStats()
{
var gaugeName = $"{m_PolicyFactory.BehaviorName}.CumulativeReward";
TimerStack.Instance.SetGauge(gaugeName, GetCumulativeReward());
}
/// <summary>
/// Sets the done flag to true and resets the agent.
/// </summary>
/// <remarks>
/// This should be used when the episode can no longer continue, such as when the Agent
/// reaches the goal or fails at the task.
/// </remarks>
/// <seealso cref="OnEpisodeBegin"/>
/// <seealso cref="EpisodeInterrupted"/>
public void EndEpisode()
{
EndEpisodeAndReset(DoneReason.DoneCalled);
}
/// <summary>
/// Indicate that the episode is over but not due to the "fault" of the Agent.
/// This has the same end result as calling <see cref="EndEpisode"/>, but has a
/// slightly different effect on training.
/// </summary>
/// <remarks>
/// This should be used when the episode could continue, but has gone on for
/// a sufficient number of steps.
/// </remarks>
/// <seealso cref="OnEpisodeBegin"/>
/// <seealso cref="EndEpisode"/>
public void EpisodeInterrupted()
{
EndEpisodeAndReset(DoneReason.MaxStepReached);
}
/// <summary>
/// Internal method to end the episode and reset the Agent.
/// </summary>
/// <param name="reason"></param>
void EndEpisodeAndReset(DoneReason reason)
{
NotifyAgentDone(reason);
_AgentReset();
}
/// <summary>
/// Requests a new decision for this agent.
/// </summary>
/// <remarks>
/// Call `RequestDecision()` whenever an agent needs a decision. You often
/// want to request a decision every environment step. However, if an agent
/// cannot use the decision every step, then you can request a decision less
/// frequently.
///
/// You can add a <seealso cref="DecisionRequester"/> component to the agent's
/// [GameObject] to drive the agent's decision making. When you use this component,
/// do not call `RequestDecision()` separately.
///
/// Note that this function calls <seealso cref="RequestAction"/>; you do not need to
/// call both functions at the same time.
///
/// [GameObject]: https://docs.unity3d.com/Manual/GameObjects.html
/// </remarks>
public void RequestDecision()
{
m_RequestDecision = true;
RequestAction();
}
/// <summary>
/// Requests an action for this agent.
/// </summary>
/// <remarks>
/// Call `RequestAction()` to repeat the previous action returned by the agent's
/// most recent decision. A new decision is not requested. When you call this function,
/// the Agent instance invokes <seealso cref="IActionReceiver.OnActionReceived"/> with the
/// existing action vector.
///
/// You can use `RequestAction()` in situations where an agent must take an action
/// every update, but doesn't need to make a decision as often. For example, an
/// agent that moves through its environment might need to apply an action to keep
/// moving, but only needs to make a decision to change course or speed occasionally.
///
/// You can add a <seealso cref="DecisionRequester"/> component to the agent's
/// [GameObject] to drive the agent's decision making and action frequency. When you
/// use this component, do not call `RequestAction()` separately.
///
/// Note that <seealso cref="RequestDecision"/> calls `RequestAction()`; you do not need to
/// call both functions at the same time.
///
/// [GameObject]: https://docs.unity3d.com/Manual/GameObjects.html
/// </remarks>
public void RequestAction()
{
m_RequestAction = true;
}
/// Helper function that resets all the data structures associated with
/// the agent. Typically used when the agent is being initialized or reset
/// at the end of an episode.
void ResetData()
{
m_ActuatorManager?.ResetData();
}
/// <summary>
/// Implement `Initialize()` to perform one-time initialization or set up of the
/// Agent instance.
/// </summary>
/// <remarks>
/// `Initialize()` is called once when the agent is first enabled. If, for example,
/// the Agent object needs references to other [GameObjects] in the scene, you
/// can collect and store those references here.
///
/// Note that <seealso cref="OnEpisodeBegin"/> is called at the start of each of
/// the agent's "episodes". You can use that function for items that need to be reset
/// for each episode.
///
/// [GameObject]: https://docs.unity3d.com/Manual/GameObjects.html
/// </remarks>
public virtual void Initialize() { }
/// <summary>
/// Implement <see cref="Heuristic"/> to choose an action for this agent using a custom heuristic.
/// </summary>
/// <remarks>
/// Implement this function to provide custom decision making logic or to support manual
/// control of an agent using keyboard, mouse, game controller input, or a script.
///
/// Your heuristic implementation can use any decision making logic you specify. Assign decision
/// values to the <see cref="ActionBuffers.ContinuousActions"/> and <see cref="ActionBuffers.DiscreteActions"/>
/// arrays , passed to your function as a parameter.
/// The same array will be reused between steps. It is up to the user to initialize
/// the values on each call, for example by calling `Array.Clear(actionsOut, 0, actionsOut.Length);`.
/// Add values to the array at the same indexes as they are used in your
/// <seealso cref="IActionReceiver.OnActionReceived"/> function, which receives this array and
/// implements the corresponding agent behavior. See [Actions] for more information
/// about agent actions.
/// Note : Do not create a new float array of action in the `Heuristic()` method,
/// as this will prevent writing floats to the original action array.
///
/// An agent calls this `Heuristic()` function to make a decision when you set its behavior
/// type to <see cref="BehaviorType.HeuristicOnly"/>. The agent also calls this function if
/// you set its behavior type to <see cref="BehaviorType.Default"/> when the
/// <see cref="Academy"/> is not connected to an external training process and you do not
/// assign a trained model to the agent.
///
/// To perform imitation learning, implement manual control of the agent in the `Heuristic()`
/// function so that you can record the demonstrations required for the imitation learning
/// algorithms. (Attach a [Demonstration Recorder] component to the agent's [GameObject] to
/// record the demonstration session to a file.)
///
/// Even when you don’t plan to use heuristic decisions for an agent or imitation learning,
/// implementing a simple heuristic function can aid in debugging agent actions and interactions
/// with its environment.
///
/// [Demonstration Recorder]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/Learning-Environment-Design-Agents.md#recording-demonstrations
/// [Actions]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/Learning-Environment-Design-Agents.md#actions
/// [GameObject]: https://docs.unity3d.com/Manual/GameObjects.html
/// </remarks>
/// <example>
/// The following example illustrates a `Heuristic()` function that provides WASD-style
/// keyboard control for an agent that can move in two dimensions as well as jump. See
/// [Input Manager] for more information about the built-in Unity input functions.
/// You can also use the [Input System package], which provides a more flexible and
/// configurable input system.
/// <code>
/// public override void Heuristic(in ActionBuffers actionsOut)
/// {
/// var continuousActionsOut = actionsOut.ContinuousActions;
/// continuousActionsOut[0] = Input.GetAxis("Horizontal");
/// continuousActionsOut[1] = Input.GetKey(KeyCode.Space) ? 1.0f : 0.0f;
/// continuousActionsOut[2] = Input.GetAxis("Vertical");
/// }
/// </code>
/// [Input Manager]: https://docs.unity3d.com/Manual/class-InputManager.html
/// [Input System package]: https://docs.unity3d.com/Packages/com.unity.inputsystem@1.0/manual/index.html
/// </example>
/// <param name="actionsOut">The <see cref="ActionBuffers"/> which contain the continuous and
/// discrete action buffers to write to.</param>
/// <seealso cref="IActionReceiver.OnActionReceived"/>
public virtual void Heuristic(in ActionBuffers actionsOut)
{
// Disable deprecation warnings so we can call the legacy overload.
#pragma warning disable CS0618
// The default implementation of Heuristic calls the
// obsolete version for backward compatibility
switch (m_PolicyFactory.BrainParameters.VectorActionSpaceType)
{
case SpaceType.Continuous:
Heuristic(m_LegacyHeuristicCache);
Array.Copy(m_LegacyHeuristicCache, actionsOut.ContinuousActions.Array, m_LegacyActionCache.Length);
actionsOut.DiscreteActions.Clear();
break;
case SpaceType.Discrete:
Heuristic(m_LegacyHeuristicCache);
var discreteActionSegment = actionsOut.DiscreteActions;
for (var i = 0; i < actionsOut.DiscreteActions.Length; i++)
{
discreteActionSegment[i] = (int)m_LegacyHeuristicCache[i];
}
actionsOut.ContinuousActions.Clear();
break;
}
#pragma warning restore CS0618
}
/// <summary>
/// Set up the list of ISensors on the Agent. By default, this will select any
/// SensorComponent's attached to the Agent.
/// </summary>
internal void InitializeSensors()
{
if (m_PolicyFactory == null)
{
m_PolicyFactory = GetComponent<BehaviorParameters>();
}
if (m_PolicyFactory.ObservableAttributeHandling != ObservableAttributeOptions.Ignore)
{
var excludeInherited =
m_PolicyFactory.ObservableAttributeHandling == ObservableAttributeOptions.ExcludeInherited;
using (TimerStack.Instance.Scoped("CreateObservableSensors"))
{
var observableSensors = ObservableAttribute.CreateObservableSensors(this, excludeInherited);
sensors.AddRange(observableSensors);
}
}
// Get all attached sensor components
SensorComponent[] attachedSensorComponents;
if (m_PolicyFactory.UseChildSensors)
{
attachedSensorComponents = GetComponentsInChildren<SensorComponent>();
}
else
{
attachedSensorComponents = GetComponents<SensorComponent>();
}
sensors.Capacity += attachedSensorComponents.Length;
foreach (var component in attachedSensorComponents)
{
sensors.Add(component.CreateSensor());
}
// Support legacy CollectObservations
var param = m_PolicyFactory.BrainParameters;
if (param.VectorObservationSize > 0)
{
collectObservationsSensor = new VectorSensor(param.VectorObservationSize);
if (param.NumStackedVectorObservations > 1)
{
var stackingSensor = new StackingSensor(
collectObservationsSensor, param.NumStackedVectorObservations);
sensors.Add(stackingSensor);
}
else
{
sensors.Add(collectObservationsSensor);
}
}
// Sort the Sensors by name to ensure determinism
sensors.Sort((x, y) => x.GetName().CompareTo(y.GetName()));
#if DEBUG
// Make sure the names are actually unique
for (var i = 0; i < sensors.Count - 1; i++)
{
Debug.Assert(
!sensors[i].GetName().Equals(sensors[i + 1].GetName()),
"Sensor names must be unique.");
}
#endif
}
void InitializeActuators()
{
ActuatorComponent[] attachedActuators;
if (m_PolicyFactory.UseChildActuators)
{
attachedActuators = GetComponentsInChildren<ActuatorComponent>();
}
else
{
attachedActuators = GetComponents<ActuatorComponent>();
}
// Support legacy OnActionReceived
// TODO don't set this up if the sizes are 0?
var param = m_PolicyFactory.BrainParameters;
m_VectorActuator = new VectorActuator(this, this, param.ActionSpec);
m_ActuatorManager = new ActuatorManager(attachedActuators.Length + 1);
m_LegacyActionCache = new float[m_VectorActuator.TotalNumberOfActions()];
m_LegacyHeuristicCache = new float[m_VectorActuator.TotalNumberOfActions()];
m_ActuatorManager.Add(m_VectorActuator);
foreach (var actuatorComponent in attachedActuators)
{
m_ActuatorManager.AddActuators(actuatorComponent.CreateActuators());
}
}
/// <summary>
/// Sends the Agent info to the linked Brain.
/// </summary>
void SendInfoToBrain()
{
if (!m_Initialized)
{
throw new UnityAgentsException("Call to SendInfoToBrain when Agent hasn't been initialized." +
"Please ensure that you are calling 'base.OnEnable()' if you have overridden OnEnable.");
}
if (m_Brain == null)
{
return;
}
if (m_Info.done)
{
m_Info.ClearActions();
}
else
{
m_Info.CopyActions(m_ActuatorManager.StoredActions);
}
UpdateSensors();
using (TimerStack.Instance.Scoped("CollectObservations"))
{
using (m_CollectObservationsChecker.Start())
{
CollectObservations(collectObservationsSensor);
}
}
using (TimerStack.Instance.Scoped("WriteActionMask"))
{
m_ActuatorManager.WriteActionMask();
}
m_Info.discreteActionMasks = m_ActuatorManager.DiscreteActionMask?.GetMask();
m_Info.reward = m_Reward;
m_Info.done = false;
m_Info.maxStepReached = false;
m_Info.episodeId = m_EpisodeId;
using (TimerStack.Instance.Scoped("RequestDecision"))
{
m_Brain.RequestDecision(m_Info, sensors);
}
// If we have any DemonstrationWriters, write the AgentInfo and sensors to them.
foreach (var demoWriter in DemonstrationWriters)
{
demoWriter.Record(m_Info, sensors);
}
}
void UpdateSensors()
{
foreach (var sensor in sensors)
{
sensor.Update();
}
}
void ResetSensors()
{
foreach (var sensor in sensors)
{
sensor.Reset();
}
}
/// <summary>
/// Implement `CollectObservations()` to collect the vector observations of
/// the agent for the step. The agent observation describes the current
/// environment from the perspective of the agent.
/// </summary>
/// <param name="sensor">
/// The vector observations for the agent.
/// </param>
/// <remarks>
/// An agent's observation is any environment information that helps
/// the agent achieve its goal. For example, for a fighting agent, its
/// observation could include distances to friends or enemies, or the
/// current level of ammunition at its disposal.
///
/// You can use a combination of vector, visual, and raycast observations for an
/// agent. If you only use visual or raycast observations, you do not need to
/// implement a `CollectObservations()` function.
///
/// Add vector observations to the <paramref name="sensor"/> parameter passed to
/// this method by calling the <seealso cref="VectorSensor"/> helper methods:
/// - <see cref="VectorSensor.AddObservation(int)"/>
/// - <see cref="VectorSensor.AddObservation(float)"/>
/// - <see cref="VectorSensor.AddObservation(Vector3)"/>
/// - <see cref="VectorSensor.AddObservation(Vector2)"/>
/// - <see cref="VectorSensor.AddObservation(Quaternion)"/>
/// - <see cref="VectorSensor.AddObservation(bool)"/>
/// - <see cref="VectorSensor.AddObservation(IEnumerable{float})"/>
/// - <see cref="VectorSensor.AddOneHotObservation(int, int)"/>
///
/// You can use any combination of these helper functions to build the agent's
/// vector of observations. You must build the vector in the same order
/// each time `CollectObservations()` is called and the length of the vector
/// must always be the same. In addition, the length of the observation must
/// match the <see cref="BrainParameters.VectorObservationSize"/>
/// attribute of the linked Brain, which is set in the Editor on the
/// **Behavior Parameters** component attached to the agent's [GameObject].
///
/// For more information about observations, see [Observations and Sensors].
///
/// [GameObject]: https://docs.unity3d.com/Manual/GameObjects.html
/// [Observations and Sensors]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/Learning-Environment-Design-Agents.md#observations-and-sensors
/// </remarks>
public virtual void CollectObservations(VectorSensor sensor)
{
}
/// <summary>
/// Returns a read-only view of the observations that were generated in
/// <see cref="CollectObservations(VectorSensor)"/>. This is mainly useful inside of a
/// <see cref="Heuristic(in ActionBuffers)"/> method to avoid recomputing the observations.
/// </summary>
/// <returns>A read-only view of the observations list.</returns>
public ReadOnlyCollection<float> GetObservations()
{
return collectObservationsSensor.GetObservations();
}
/// <summary>
/// Implement `WriteDiscreteActionMask()` to collects the masks for discrete
/// actions. When using discrete actions, the agent will not perform the masked
/// action.
/// </summary>
/// <param name="actionMask">
/// The action mask for the agent.
/// </param>
/// <remarks>
/// When using Discrete Control, you can prevent the Agent from using a certain
/// action by masking it with <see cref="IDiscreteActionMask.WriteMask(int, IEnumerable{int})"/>.
///
/// See [Agents - Actions] for more information on masking actions.
///
/// [Agents - Actions]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/Learning-Environment-Design-Agents.md#actions
/// </remarks>
/// <seealso cref="IActionReceiver.OnActionReceived"/>
public virtual void WriteDiscreteActionMask(IDiscreteActionMask actionMask)
{
if (m_ActionMasker == null)
{
m_ActionMasker = new DiscreteActionMasker(actionMask);
}
// Disable deprecation warnings so we can call the legacy overload.
#pragma warning disable CS0618
CollectDiscreteActionMasks(m_ActionMasker);
#pragma warning restore CS0618
}
/// <summary>
/// Implement `OnActionReceived()` to specify agent behavior at every step, based
/// on the provided action.
/// </summary>
/// <remarks>
/// An action is passed to this function in the form of an <seealso cref="ActionBuffers"/>.
/// Your implementation must use the array to direct the agent's behavior for the
/// current step.
///
/// You decide how many elements you need in the ActionBuffers to control your
/// agent and what each element means. For example, if you want to apply a
/// force to move an agent around the environment, you can arbitrarily pick
/// three values in ActionBuffers.ContinuousActions array to use as the force components.
/// During training, the agent's policy learns to set those particular elements of
/// the array to maximize the training rewards the agent receives. (Of course,
/// if you implement a <seealso cref="Agent.Heuristic(in ActionBuffers)"/> function, it must use the same
/// elements of the action array for the same purpose since there is no learning
/// involved.)
///
/// An Agent can use continuous and/or discrete actions. Configure this along with the size
/// of the action array, in the <see cref="BrainParameters"/> of the agent's associated
/// <see cref="BehaviorParameters"/> component.
///
/// When an agent uses continuous actions, the values in the ActionBuffers.ContinuousActions
/// array are floating point numbers. You should clamp the values to the range,
/// -1..1, to increase numerical stability during training.
///
/// When an agent uses discrete actions, the values in the ActionBuffers.DiscreteActions array
/// are integers that each represent a specific, discrete action. For example,
/// you could define a set of discrete actions such as:
///
/// <code>
/// 0 = Do nothing
/// 1 = Move one space left
/// 2 = Move one space right
/// 3 = Move one space up
/// 4 = Move one space down
/// </code>
///
/// When making a decision, the agent picks one of the five actions and puts the
/// corresponding integer value in the ActionBuffers.DiscreteActions array. For example, if the agent
/// decided to move left, the ActionBuffers.DiscreteActions parameter would be an array with
/// a single element with the value 1.
///
/// You can define multiple sets, or branches, of discrete actions to allow an
/// agent to perform simultaneous, independent actions. For example, you could
/// use one branch for movement and another branch for throwing a ball left, right,
/// up, or down, to allow the agent to do both in the same step.
///
/// The ActionBuffers.DiscreteActions array of an agent with discrete actions contains one
/// element for each branch. The value of each element is the integer representing the
/// chosen action for that branch. The agent always chooses one action for each branch.
///
/// When you use the discrete actions, you can prevent the training process
/// or the neural network model from choosing specific actions in a step by
/// implementing the <see cref="WriteDiscreteActionMask(IDiscreteActionMask)"/>
/// method. For example, if your agent is next to a wall, you could mask out any
/// actions that would result in the agent trying to move into the wall.
///
/// For more information about implementing agent actions see [Agents - Actions].
///
/// [Agents - Actions]: https://github.com/Unity-Technologies/ml-agents/blob/release_12_docs/docs/Learning-Environment-Design-Agents.md#actions
/// </remarks>
/// <param name="actions">
/// Struct containing the buffers of actions to be executed at this step.
/// </param>
public virtual void OnActionReceived(ActionBuffers actions)
{
var actionSpec = m_PolicyFactory.BrainParameters.ActionSpec;
// For continuous and discrete actions together, we don't need to fall back to the legacy method
if (actionSpec.NumContinuousActions > 0 && actionSpec.NumDiscreteActions > 0)
{
// Nothing implemented.
return;
}
if (!actions.ContinuousActions.IsEmpty())
{
Array.Copy(actions.ContinuousActions.Array,
m_LegacyActionCache,
actionSpec.NumContinuousActions);
}
else
{
for (var i = 0; i < m_LegacyActionCache.Length; i++)
{
m_LegacyActionCache[i] = (float)actions.DiscreteActions[i];
}
}
// Disable deprecation warnings so we can call the legacy overload.
#pragma warning disable CS0618
OnActionReceived(m_LegacyActionCache);
#pragma warning restore CS0618
}
/// <summary>
/// Implement `OnEpisodeBegin()` to set up an Agent instance at the beginning
/// of an episode.
/// </summary>
/// <seealso cref="Initialize"/>
/// <seealso cref="EndEpisode"/>
public virtual void OnEpisodeBegin() { }
/// <summary>
/// Gets the most recent ActionBuffer for this agent.
/// </summary>
/// <returns>The most recent ActionBuffer for this agent</returns>
public ActionBuffers GetStoredActionBuffers()
{
return m_ActuatorManager.StoredActions;
}
/// <summary>
/// An internal reset method that updates internal data structures in
/// addition to calling <see cref="OnEpisodeBegin"/>.
/// </summary>
void _AgentReset()
{
ResetData();
m_StepCount = 0;
using (m_OnEpisodeBeginChecker.Start())
{
OnEpisodeBegin();
}
}
/// <summary>
/// Scales continuous action from [-1, 1] to arbitrary range.
/// </summary>
/// <param name="rawAction">The input action value.</param>
/// <param name="min">The minimum output value.</param>
/// <param name="max">The maximum output value.</param>
/// <returns>The <paramref name="rawAction"/> scaled from [-1,1] to
/// [<paramref name="min"/>, <paramref name="max"/>].</returns>
protected static float ScaleAction(float rawAction, float min, float max)
{
var middle = (min + max) / 2;
var range = (max - min) / 2;
return rawAction * range + middle;
}
/// <summary>
/// Signals the agent that it must send its decision to the brain.
/// </summary>
void SendInfo()
{
// If the Agent is done, it has just reset and thus requires a new decision
if (m_RequestDecision)
{
SendInfoToBrain();
m_Reward = 0f;
m_RequestDecision = false;
}
}
void AgentIncrementStep()
{
m_StepCount += 1;
}
/// Used by the brain to make the agent perform a step.
void AgentStep()
{
if ((m_RequestAction) && (m_Brain != null))
{
m_RequestAction = false;
m_ActuatorManager.ExecuteActions();
}
if ((m_StepCount >= MaxStep) && (MaxStep > 0))
{
NotifyAgentDone(DoneReason.MaxStepReached);
_AgentReset();
}
}
void DecideAction()
{
if (m_ActuatorManager.StoredActions.ContinuousActions.Array == null)
{
ResetData();
}
var actions = m_Brain?.DecideAction() ?? new ActionBuffers();
m_Info.CopyActions(actions);
m_ActuatorManager.UpdateActions(actions);
}
}
}