using Unity.Barracuda;
using System.Collections.Generic;
using System.Diagnostics;
using Unity.MLAgents.Actuators;
using Unity.MLAgents.Inference;
using Unity.MLAgents.Sensors;
namespace Unity.MLAgents.Policies
{
///
/// Where to perform inference.
///
public enum InferenceDevice
{
///
/// Default inference. This is currently the same as Burst, but may change in the future.
///
Default = 0,
///
/// GPU inference. Corresponds to WorkerFactory.Type.ComputePrecompiled in Barracuda.
///
GPU = 1,
///
/// CPU inference using Burst. Corresponds to WorkerFactory.Type.CSharpBurst in Barracuda.
///
Burst = 2,
///
/// CPU inference. Corresponds to in WorkerFactory.Type.CSharp Barracuda.
/// Burst is recommended instead; this is kept for legacy compatibility.
///
CPU = 3,
}
///
/// The Barracuda Policy uses a Barracuda Model to make decisions at
/// every step. It uses a ModelRunner that is shared across all
/// Barracuda Policies that use the same model and inference devices.
///
internal class BarracudaPolicy : IPolicy
{
protected ModelRunner m_ModelRunner;
ActionBuffers m_LastActionBuffer;
int m_AgentId;
///
/// Sensor shapes for the associated Agents. All Agents must have the same shapes for their Sensors.
///
List m_SensorShapes;
ActionSpec m_ActionSpec;
private string m_BehaviorName;
///
/// List of actuators, only used for analytics
///
private IList m_Actuators;
///
/// Whether or not we've tried to send analytics for this model. We only ever try to send once per policy,
/// and do additional deduplication in the analytics code.
///
private bool m_AnalyticsSent;
///
/// Instantiate a BarracudaPolicy with the necessary objects for it to run.
///
/// The action spec of the behavior.
/// The actuators used for this behavior.
/// The Neural Network to use.
/// Which device Barracuda will run on.
/// The name of the behavior.
public BarracudaPolicy(
ActionSpec actionSpec,
IList actuators,
NNModel model,
InferenceDevice inferenceDevice,
string behaviorName
)
{
var modelRunner = Academy.Instance.GetOrCreateModelRunner(model, actionSpec, inferenceDevice);
m_ModelRunner = modelRunner;
m_BehaviorName = behaviorName;
m_ActionSpec = actionSpec;
m_Actuators = actuators;
}
///
public void RequestDecision(AgentInfo info, List sensors)
{
SendAnalytics(sensors);
m_AgentId = info.episodeId;
m_ModelRunner?.PutObservations(info, sensors);
}
[Conditional("MLA_UNITY_ANALYTICS_MODULE")]
void SendAnalytics(IList sensors)
{
if (!m_AnalyticsSent)
{
m_AnalyticsSent = true;
Analytics.InferenceAnalytics.InferenceModelSet(
m_ModelRunner.Model,
m_BehaviorName,
m_ModelRunner.InferenceDevice,
sensors,
m_ActionSpec,
m_Actuators
);
}
}
///
public ref readonly ActionBuffers DecideAction()
{
if (m_ModelRunner == null)
{
m_LastActionBuffer = ActionBuffers.Empty;
}
else
{
m_ModelRunner?.DecideBatch();
m_LastActionBuffer = m_ModelRunner.GetAction(m_AgentId);
}
return ref m_LastActionBuffer;
}
public void Dispose()
{
}
}
}