Merge solver code with static interface

- Breaks backwards compat solver code with last version. Concurrent broke backwards compat because of race conditions with rng, but single is now broken too, despite it being 2x faster (!!!!)
- Literally twice as fast as Rust now in single thread
- Concurrent doesn't work yet, deadlocks somewhere..?
This commit is contained in:
Asriel Camora
2023-07-07 09:58:47 +02:00
parent 2894867195
commit d5a8288439
9 changed files with 332 additions and 145 deletions
+2 -2
View File
@@ -55,10 +55,10 @@ internal static class Program
Debugger.Break(); Debugger.Break();
var s = Stopwatch.StartNew(); var s = Stopwatch.StartNew();
if (true) if (true)
_ = Solver.Crafty.Solver.SearchStepwise(config, input, a => Console.WriteLine(a)); _ = SolverUtils.SearchStepwise<SolverSingle>(config, input, a => Console.WriteLine(a));
else else
{ {
(var actions, _) = Solver.Crafty.Solver.SearchOneshot(config, input); (var actions, _) = SolverUtils.SearchOneshot<SolverConcurrent>(config, input);
foreach (var action in actions) foreach (var action in actions)
Console.Write($">{action.IntName()}"); Console.Write($">{action.IntName()}");
Console.WriteLine(); Console.WriteLine();
+2 -1
View File
@@ -1,5 +1,6 @@
using Craftimizer.Simulator; using Craftimizer.Simulator;
using Craftimizer.Simulator.Actions; using Craftimizer.Simulator.Actions;
using Craftimizer.Solver.Crafty;
using Dalamud.Interface.Windowing; using Dalamud.Interface.Windowing;
using System; using System;
using System.Collections.Concurrent; using System.Collections.Concurrent;
@@ -71,7 +72,7 @@ public sealed partial class SimulatorWindow : Window, IDisposable
SolverInitialActionCount = Actions.Count; SolverInitialActionCount = Actions.Count;
SolverTaskToken = new(); SolverTaskToken = new();
SolverTask = Task.Run(() => Solver.Crafty.Solver.SearchStepwise(Service.Configuration.SolverConfig, solverState, SolverActionQueue.Enqueue, SolverTaskToken.Token)); SolverTask = Task.Run(() => SolverUtils.SearchStepwise<SolverConcurrent>(Service.Configuration.SolverConfig, solverState, SolverActionQueue.Enqueue, SolverTaskToken.Token));
} }
public void Dispose() public void Dispose()
+43 -6
View File
@@ -17,11 +17,11 @@ public struct ActionSet
private static ActionType ToAction(int index) => Simulator.AcceptedActions[index]; private static ActionType ToAction(int index) => Simulator.AcceptedActions[index];
[Pure] [Pure]
[MethodImpl(MethodImplOptions.AggressiveInlining)] [MethodImpl(MethodImplOptions.AggressiveInlining)]
private static uint ToMask(ActionType action) => 1u << (FromAction(action) + 1); private static uint ToMask(ActionType action) => 1u << FromAction(action) + 1;
// Return true if action was newly added and not there before. // Return true if action was newly added and not there before.
[MethodImpl(MethodImplOptions.AggressiveInlining)] [MethodImpl(MethodImplOptions.AggressiveInlining)]
public bool AddAction(ActionType action) public bool AddActionConcurrent(ActionType action)
{ {
var mask = ToMask(action); var mask = ToMask(action);
var old = Interlocked.Or(ref bits, mask); var old = Interlocked.Or(ref bits, mask);
@@ -30,13 +30,33 @@ public struct ActionSet
// Return true if action was newly removed and not already gone. // Return true if action was newly removed and not already gone.
[MethodImpl(MethodImplOptions.AggressiveInlining)] [MethodImpl(MethodImplOptions.AggressiveInlining)]
public bool RemoveAction(ActionType action) public bool RemoveActionConcurrent(ActionType action)
{ {
var mask = ToMask(action); var mask = ToMask(action);
var old = Interlocked.And(ref bits, ~mask); var old = Interlocked.And(ref bits, ~mask);
return (old & mask) != 0; return (old & mask) != 0;
} }
// Return true if action was newly added and not there before.
[MethodImpl(MethodImplOptions.AggressiveInlining)]
public bool AddAction(ActionType action)
{
var mask = ToMask(action);
var old = bits;
bits |= mask;
return (old & mask) == 0;
}
// Return true if action was newly removed and not already gone.
[MethodImpl(MethodImplOptions.AggressiveInlining)]
public bool RemoveAction(ActionType action)
{
var mask = ToMask(action);
var old = bits;
bits &= ~mask;
return (old & mask) != 0;
}
[Pure] [Pure]
[MethodImpl(MethodImplOptions.AggressiveInlining)] [MethodImpl(MethodImplOptions.AggressiveInlining)]
public readonly bool HasAction(ActionType action) => (bits & ToMask(action)) != 0; public readonly bool HasAction(ActionType action) => (bits & ToMask(action)) != 0;
@@ -51,10 +71,10 @@ public struct ActionSet
public readonly bool IsEmpty => bits == 0; public readonly bool IsEmpty => bits == 0;
[MethodImpl(MethodImplOptions.AggressiveInlining)] [MethodImpl(MethodImplOptions.AggressiveInlining)]
public readonly ActionType SelectRandom(Random random) => ElementAt(random.Next(Count)); public readonly ActionType SelectRandom(Random random) => ElementAt(0);// random.Next(Count));
[MethodImpl(MethodImplOptions.AggressiveInlining)] [MethodImpl(MethodImplOptions.AggressiveInlining)]
public ActionType? PopRandom(Random random) public ActionType? PopRandomConcurrent(Random random)
{ {
uint snapshot; uint snapshot;
uint newValue; uint newValue;
@@ -76,7 +96,7 @@ public struct ActionSet
} }
[MethodImpl(MethodImplOptions.AggressiveInlining)] [MethodImpl(MethodImplOptions.AggressiveInlining)]
public ActionType? PopFirst() public ActionType? PopFirstConcurrent()
{ {
uint snapshot; uint snapshot;
uint newValue; uint newValue;
@@ -94,6 +114,23 @@ public struct ActionSet
return action; return action;
} }
[MethodImpl(MethodImplOptions.AggressiveInlining)]
public ActionType PopRandom(Random random)
{
return PopFirst();
var action = ElementAt(random.Next(Count));
RemoveAction(action);
return action;
}
[MethodImpl(MethodImplOptions.AggressiveInlining)]
public ActionType PopFirst()
{
var action = First();
RemoveAction(action);
return action;
}
[Pure] [Pure]
[MethodImpl(MethodImplOptions.AggressiveInlining)] [MethodImpl(MethodImplOptions.AggressiveInlining)]
public readonly ActionType First() => ElementAt(0); public readonly ActionType First() => ElementAt(0);
+24 -3
View File
@@ -20,17 +20,17 @@ public sealed class ArenaNode<T> where T : struct
private static int BatchCount = MaxSize / BatchSize; private static int BatchCount = MaxSize / BatchSize;
public ArenaNode<T>[][] Data; public ArenaNode<T>[][] Data;
private int index; private int index; // Unused in single threaded workload
private int count; private int count;
public readonly int Count => count; public readonly int Count => count;
public void Add(ArenaNode<T> node) public void AddConcurrent(ArenaNode<T> node)
{ {
if (Data == null) if (Data == null)
Interlocked.CompareExchange(ref Data, new ArenaNode<T>[BatchCount][], null); Interlocked.CompareExchange(ref Data, new ArenaNode<T>[BatchCount][], null);
var idx = Interlocked.Increment(ref this.index) - 1; var idx = Interlocked.Increment(ref index) - 1;
var (arrayIdx, subIdx) = GetArrayIndex(idx); var (arrayIdx, subIdx) = GetArrayIndex(idx);
@@ -42,6 +42,19 @@ public sealed class ArenaNode<T> where T : struct
Interlocked.Increment(ref count); Interlocked.Increment(ref count);
} }
public void Add(ArenaNode<T> node)
{
Data ??= new ArenaNode<T>[BatchCount][];
var idx = count++;
var (arrayIdx, subIdx) = GetArrayIndex(idx);
Data[arrayIdx] ??= new ArenaNode<T>[BatchSize];
Data[arrayIdx][subIdx] = node;
}
[Pure] [Pure]
[MethodImpl(MethodImplOptions.AggressiveInlining)] [MethodImpl(MethodImplOptions.AggressiveInlining)]
private static (int arrayIdx, int subIdx) GetArrayIndex(int idx) => private static (int arrayIdx, int subIdx) GetArrayIndex(int idx) =>
@@ -59,6 +72,14 @@ public sealed class ArenaNode<T> where T : struct
Parent = parent; Parent = parent;
} }
[MethodImpl(MethodImplOptions.AggressiveInlining)]
public ArenaNode<T> AddConcurrent(T state)
{
var node = new ArenaNode<T>(state, this);
Children.AddConcurrent(node);
return node;
}
[MethodImpl(MethodImplOptions.AggressiveInlining)] [MethodImpl(MethodImplOptions.AggressiveInlining)]
public ArenaNode<T> Add(T state) public ArenaNode<T> Add(T state)
{ {
+12
View File
@@ -0,0 +1,12 @@
using Node = Craftimizer.Solver.Crafty.ArenaNode<Craftimizer.Solver.Crafty.SimulationNode>;
namespace Craftimizer.Solver.Crafty;
public interface ISolver
{
abstract static void LoadChildData(Span<float> scoreSums, Span<int> visits, Span<float> maxScores, ref Node[] chunk, int iterCount);
abstract static bool SearchIter(ref SolverConfig config, Node rootNode, Random random, Simulator simulator);
abstract static void Search(ref SolverConfig config, Node rootNode, CancellationToken token);
}
+8 -1
View File
@@ -9,10 +9,17 @@ public struct NodeScores
public float MaxScore; public float MaxScore;
public int Visits; public int Visits;
public void Visit(float score) public void VisitConcurrent(float score)
{ {
Intrinsics.CASAdd(ref ScoreSum, score); Intrinsics.CASAdd(ref ScoreSum, score);
Intrinsics.CASMax(ref MaxScore, score); Intrinsics.CASMax(ref MaxScore, score);
Interlocked.Increment(ref Visits); Interlocked.Increment(ref Visits);
} }
public void Visit(float score)
{
ScoreSum += score;
MaxScore = Math.Max(MaxScore, score);
Visits++;
}
} }
+97
View File
@@ -0,0 +1,97 @@
using System.Diagnostics.Contracts;
using System.Runtime.CompilerServices;
using Node = Craftimizer.Solver.Crafty.ArenaNode<Craftimizer.Solver.Crafty.SimulationNode>;
namespace Craftimizer.Solver.Crafty;
// https://github.com/alostsock/crafty/blob/cffbd0cad8bab3cef9f52a3e3d5da4f5e3781842/crafty/src/simulator.rs
public sealed class SolverConcurrent : ISolver
{
[MethodImpl(MethodImplOptions.AggressiveInlining | MethodImplOptions.AggressiveOptimization)]
public static void LoadChildData(Span<float> scoreSums, Span<int> visits, Span<float> maxScores, ref Node[] chunk, int iterCount)
{
for (var j = 0; j < iterCount; ++j)
{
var node = chunk[j]?.State.Scores ?? new();
scoreSums[j] = node.ScoreSum;
visits[j] = node.Visits;
maxScores[j] = node.MaxScore;
}
}
[Pure]
[MethodImpl(MethodImplOptions.AggressiveInlining)]
public static Node? EvalBestChild(ref SolverConfig config, int parentVisits, ref Node.ChildBuffer children) =>
parentVisits == 0 ?
null :
SolverUtils.EvalBestChild<SolverConcurrent>(ref config, parentVisits, ref children);
[Pure]
public static Node Select(ref SolverConfig config, Node rootNode)
{
var node = rootNode;
while (true)
{
var expandable = !node.State.AvailableActions.IsEmpty;
var likelyTerminal = node.Children.Count == 0;
if (expandable || likelyTerminal)
return node;
// select the node with the highest score
// if null (current node is invalid & not backpropagated just yet), try again from root
node = EvalBestChild(ref config, node.State.Scores.Visits, ref node.Children) ?? rootNode;
}
}
public static (Node ExpandedNode, float Score)? ExpandAndRollout(ref SolverConfig config, Node rootNode, Random random, Simulator simulator, Node initialNode)
{
ref var initialState = ref initialNode.State;
// expand once
if (initialState.IsComplete)
return (initialNode, initialState.CalculateScore(config.MaxStepCount) ?? 0);
var poppedAction = initialState.AvailableActions.PopRandomConcurrent(random);
if (!poppedAction.HasValue)
return null;
var expandedNode = initialNode.Add(SolverUtils.Execute(simulator, initialState.State, poppedAction.Value, true));
return SolverUtils.Rollout(ref config, rootNode, expandedNode, random, simulator);
}
public static void Backpropagate(Node rootNode, Node startNode, float score)
{
while (true)
{
startNode.State.Scores.VisitConcurrent(score);
if (startNode == rootNode)
break;
startNode = startNode.Parent!;
}
}
public static bool SearchIter(ref SolverConfig config, Node rootNode, Random random, Simulator simulator)
{
var selectedNode = Select(ref config, rootNode);
var rolledOut = ExpandAndRollout(ref config, rootNode, random, simulator, selectedNode);
if (!rolledOut.HasValue)
return false;
var (endNode, score) = rolledOut.Value;
Backpropagate(rootNode, endNode, score);
return true;
}
public static void SearchThread(SolverConfig config, Node rootNode, CancellationToken token) =>
SolverUtils.Search<SolverConcurrent>(ref config, rootNode, token);
public static void Search(ref SolverConfig config, Node rootNode, CancellationToken token)
{
var configP = config;
var tasks = new Task[config.ThreadCount];
for (var i = 0; i < config.ThreadCount; ++i)
tasks[i] = Task.Run(() => SearchThread(configP, rootNode, token), token);
Task.WaitAll(tasks, CancellationToken.None);
}
}
+79
View File
@@ -0,0 +1,79 @@
using System.Diagnostics.Contracts;
using System.Runtime.CompilerServices;
using Node = Craftimizer.Solver.Crafty.ArenaNode<Craftimizer.Solver.Crafty.SimulationNode>;
namespace Craftimizer.Solver.Crafty;
// https://github.com/alostsock/crafty/blob/cffbd0cad8bab3cef9f52a3e3d5da4f5e3781842/crafty/src/simulator.rs
public sealed class SolverSingle : ISolver
{
[MethodImpl(MethodImplOptions.AggressiveInlining | MethodImplOptions.AggressiveOptimization)]
public static void LoadChildData(Span<float> scoreSums, Span<int> visits, Span<float> maxScores, ref Node[] chunk, int iterCount)
{
for (var j = 0; j < iterCount; ++j)
{
ref var node = ref chunk[j].State.Scores;
scoreSums[j] = node.ScoreSum;
visits[j] = node.Visits;
maxScores[j] = node.MaxScore;
}
}
[Pure]
[MethodImpl(MethodImplOptions.AggressiveInlining)]
public static Node EvalBestChild(ref SolverConfig config, int parentVisits, ref Node.ChildBuffer children) =>
SolverUtils.EvalBestChild<SolverSingle>(ref config, parentVisits, ref children);
[Pure]
public static Node Select(ref SolverConfig config, Node node)
{
while (true)
{
var expandable = !node.State.AvailableActions.IsEmpty;
var likelyTerminal = node.Children.Count == 0;
if (expandable || likelyTerminal)
return node;
// select the node with the highest score
node = EvalBestChild(ref config, node.State.Scores.Visits, ref node.Children);
}
}
public static (Node ExpandedNode, float Score) ExpandAndRollout(ref SolverConfig config, Node rootNode, Random random, Simulator simulator, Node initialNode)
{
ref var initialState = ref initialNode.State;
// expand once
if (initialState.IsComplete)
return (initialNode, initialState.CalculateScore(config.MaxStepCount) ?? 0);
var poppedAction = initialState.AvailableActions.PopRandom(random);
var expandedNode = initialNode.Add(SolverUtils.Execute(simulator, initialState.State, poppedAction, true));
return SolverUtils.Rollout(ref config, rootNode, expandedNode, random, simulator);
}
public static void Backpropagate(Node rootNode, Node startNode, float score)
{
while (true)
{
startNode.State.Scores.Visit(score);
if (startNode == rootNode)
break;
startNode = startNode.Parent!;
}
}
public static bool SearchIter(ref SolverConfig config, Node rootNode, Random random, Simulator simulator)
{
var selectedNode = Select(ref config, rootNode);
var (endNode, score) = ExpandAndRollout(ref config, rootNode, random, simulator, selectedNode);
Backpropagate(rootNode, endNode, score);
return true;
}
public static void Search(ref SolverConfig config, Node rootNode, CancellationToken token) =>
SolverUtils.Search<SolverSingle>(ref config, rootNode, token);
}
@@ -1,38 +1,14 @@
using Craftimizer.Simulator;
using Craftimizer.Simulator.Actions; using Craftimizer.Simulator.Actions;
using Craftimizer.Simulator;
using Node = Craftimizer.Solver.Crafty.ArenaNode<Craftimizer.Solver.Crafty.SimulationNode>;
using System.Diagnostics.Contracts; using System.Diagnostics.Contracts;
using System.Numerics; using System.Numerics;
using System.Runtime.CompilerServices; using System.Runtime.CompilerServices;
using Node = Craftimizer.Solver.Crafty.ArenaNode<Craftimizer.Solver.Crafty.SimulationNode>;
namespace Craftimizer.Solver.Crafty; namespace Craftimizer.Solver.Crafty;
public static class SolverUtils
// https://github.com/alostsock/crafty/blob/cffbd0cad8bab3cef9f52a3e3d5da4f5e3781842/crafty/src/simulator.rs
public sealed class Solver
{ {
public SolverConfig Config; public static SimulationNode Execute(Simulator simulator, SimulationState state, ActionType action, bool strict)
public Node RootNode;
public Random Random;
public Solver(SolverConfig config, SimulationState state, bool strict)
{
Config = config;
var sim = new Simulator(state, config.MaxStepCount);
RootNode = new(new(
state,
null,
sim.CompletionState,
sim.AvailableActionsHeuristic(strict)
));
Random = state.Input.Random;
}
public Solver(SolverConfig config, SimulationInput input, bool strict) : this(config, new SimulationState(input), strict)
{
}
private static SimulationNode Execute(Simulator simulator, SimulationState state, ActionType action, bool strict)
{ {
(_, var newState) = simulator.Execute(state, action); (_, var newState) = simulator.Execute(state, action);
return new( return new(
@@ -53,7 +29,7 @@ public sealed class Solver
if (!state.AvailableActions.HasAction(action)) if (!state.AvailableActions.HasAction(action))
return (startNode, CompletionState.InvalidAction); return (startNode, CompletionState.InvalidAction);
state.AvailableActions.RemoveAction(action); state.AvailableActions.RemoveActionConcurrent(action);
startNode = startNode.Add(Execute(simulator, state.State, action, strict)); startNode = startNode.Add(Execute(simulator, state.State, action, strict));
} }
@@ -63,7 +39,7 @@ public sealed class Solver
[Pure] [Pure]
[MethodImpl(MethodImplOptions.AggressiveInlining)] [MethodImpl(MethodImplOptions.AggressiveInlining)]
private static Node ChildMaxScore(ref Node.ChildBuffer children) public static Node ChildMaxScore(ref Node.ChildBuffer children)
{ {
var length = children.Count; var length = children.Count;
var vecLength = Vector<float>.Count; var vecLength = Vector<float>.Count;
@@ -95,17 +71,29 @@ public sealed class Solver
} }
[Pure] [Pure]
[MethodImpl(MethodImplOptions.AggressiveInlining)] public static (List<ActionType> Actions, SimulationNode Node) Solution(Node node)
private Node? EvalBestChild(int parentVisits, ref Node.ChildBuffer children)
{ {
if (parentVisits == 0) var actions = new List<ActionType>();
return null; while (node.Children.Count != 0)
{
node = ChildMaxScore(ref node.Children);
if (node.State.Action != null)
actions.Add(node.State.Action.Value);
}
return (actions, node.State);
}
[Pure]
[MethodImpl(MethodImplOptions.AggressiveInlining | MethodImplOptions.AggressiveOptimization)]
public static Node EvalBestChild<S>(ref SolverConfig config, int parentVisits, ref Node.ChildBuffer children) where S : ISolver
{
var length = children.Count; var length = children.Count;
var vecLength = Vector<float>.Count; var vecLength = Vector<float>.Count;
var C = MathF.Sqrt(Config.ExplorationConstant * MathF.Log(parentVisits)); var C = MathF.Sqrt(config.ExplorationConstant * MathF.Log(parentVisits));
var w = Config.MaxScoreWeightingConstant; var w = config.MaxScoreWeightingConstant;
var W = 1f - w; var W = 1f - w;
var CVector = new Vector<float>(C); var CVector = new Vector<float>(C);
@@ -119,14 +107,7 @@ public sealed class Solver
{ {
var iterCount = Math.Min(vecLength, length); var iterCount = Math.Min(vecLength, length);
ref var chunk = ref children.Data[i]; S.LoadChildData(scoreSums, visits, maxScores, ref children.Data[i], iterCount);
for (var j = 0; j < iterCount; ++j)
{
var node = chunk[j]?.State.Scores ?? new();
scoreSums[j] = node.ScoreSum;
visits[j] = node.Visits;
maxScores[j] = node.MaxScore;
}
var s = new Vector<float>(scoreSums); var s = new Vector<float>(scoreSums);
var m = new Vector<float>(maxScores); var m = new Vector<float>(maxScores);
@@ -151,47 +132,21 @@ public sealed class Solver
return children.Data[max.Item1][max.Item2]; return children.Data[max.Item1][max.Item2];
} }
[Pure] [MethodImpl(MethodImplOptions.AggressiveInlining)]
public Node Select() public static (Node ExpandedNode, float Score) Rollout(ref SolverConfig config, Node rootNode, Node expandedNode, Random random, Simulator simulator)
{ {
var node = RootNode;
while (true)
{
var expandable = !node.State.AvailableActions.IsEmpty;
var likelyTerminal = node.Children.Count == 0;
if (expandable || likelyTerminal)
return node;
// select the node with the highest score
// if null (current node is invalid & not backpropagated just yet), try again from root
node = EvalBestChild(node.State.Scores.Visits, ref node.Children) ?? RootNode;
}
}
public (Node ExpandedNode, float Score)? ExpandAndRollout(Simulator simulator, Node initialNode)
{
ref var initialState = ref initialNode.State;
// expand once
if (initialState.IsComplete)
return (initialNode, initialState.CalculateScore(Config.MaxStepCount) ?? 0);
var poppedAction = initialState.AvailableActions.PopRandom(Random);
if (!poppedAction.HasValue)
return null;
var expandedNode = initialNode.Add(Execute(simulator, initialState.State, poppedAction.Value, true));
// playout to a terminal state // playout to a terminal state
var currentState = expandedNode.State.State; var currentState = expandedNode.State.State;
var currentCompletionState = expandedNode.State.SimulationCompletionState; var currentCompletionState = expandedNode.State.SimulationCompletionState;
var currentActions = expandedNode.State.AvailableActions; var currentActions = expandedNode.State.AvailableActions;
byte actionCount = 0; byte actionCount = 0;
Span<ActionType> actions = stackalloc ActionType[Config.MaxStepCount]; Span<ActionType> actions = stackalloc ActionType[config.MaxStepCount - currentState.ActionCount];
while (true) while (true)
{ {
if (SimulationNode.GetCompletionState(currentCompletionState, currentActions) != CompletionState.Incomplete) if (SimulationNode.GetCompletionState(currentCompletionState, currentActions) != CompletionState.Incomplete)
break; break;
var nextAction = currentActions.SelectRandom(Random); var nextAction = currentActions.SelectRandom(random);
actions[actionCount++] = nextAction; actions[actionCount++] = nextAction;
(_, currentState) = simulator.Execute(currentState, nextAction); (_, currentState) = simulator.Execute(currentState, nextAction);
currentCompletionState = simulator.CompletionState; currentCompletionState = simulator.CompletionState;
@@ -199,10 +154,10 @@ public sealed class Solver
} }
// store the result if a max score was reached // store the result if a max score was reached
var score = SimulationNode.CalculateScoreForState(currentState, currentCompletionState, Config.MaxStepCount) ?? 0; var score = SimulationNode.CalculateScoreForState(currentState, currentCompletionState, config.MaxStepCount) ?? 0;
if (currentCompletionState == CompletionState.ProgressComplete) if (currentCompletionState == CompletionState.ProgressComplete)
{ {
if (score >= Config.ScoreStorageThreshold && score >= RootNode.State.Scores.MaxScore) if (score >= config.ScoreStorageThreshold && score >= rootNode.State.Scores.MaxScore)
{ {
(var terminalNode, _) = ExecuteActions(simulator, expandedNode, actions[..actionCount], true); (var terminalNode, _) = ExecuteActions(simulator, expandedNode, actions[..actionCount], true);
return (terminalNode, score); return (terminalNode, score);
@@ -211,80 +166,58 @@ public sealed class Solver
return (expandedNode, score); return (expandedNode, score);
} }
public void Backpropagate(Node startNode, float score) [MethodImpl(MethodImplOptions.AggressiveInlining)]
public static void Search<S>(ref SolverConfig config, Node rootNode, CancellationToken token) where S : ISolver
{ {
while (true) Simulator simulator = new(rootNode.State.State, config.MaxStepCount);
{ var random = rootNode.State.State.Input.Random;
startNode.State.Scores.Visit(score); for (var i = 0; i < config.Iterations; i++)
if (startNode == RootNode)
break;
startNode = startNode.Parent!;
}
}
public void SearchThread(CancellationToken token)
{
Simulator simulator = new(RootNode.State.State, Config.MaxStepCount);
for (var i = 0; i < Config.Iterations; i++)
{ {
if (token.IsCancellationRequested) if (token.IsCancellationRequested)
break; break;
var selectedNode = Select(); if (!S.SearchIter(ref config, rootNode, random, simulator))
var rolledOut = ExpandAndRollout(simulator, selectedNode);
if (!rolledOut.HasValue)
{ {
// Retry, count this iteration as moot // Retry, count this iteration as moot
i--; i--;
continue; continue;
} }
var (endNode, score) = rolledOut.Value;
Backpropagate(endNode, score);
} }
} }
public void Search(CancellationToken token)
{
var tasks = new Task[Config.ThreadCount];
for (var i = 0; i < Config.ThreadCount; ++i)
tasks[i] = Task.Run(() => SearchThread(token), token);
Task.WaitAll(tasks, CancellationToken.None);
}
[Pure] [Pure]
public (List<ActionType> Actions, SimulationNode Node) Solution() [MethodImpl(MethodImplOptions.AggressiveInlining)]
public static Node CreateRootNode(SolverConfig config, SimulationInput input, bool strict) =>
CreateRootNode(config, new SimulationState(input), strict);
[Pure]
[MethodImpl(MethodImplOptions.AggressiveInlining)]
public static Node CreateRootNode(SolverConfig config, SimulationState state, bool strict)
{ {
var actions = new List<ActionType>(); var sim = new Simulator(state, config.MaxStepCount);
var node = RootNode; return new(new(
while (node.Children.Count != 0) state,
{ null,
node = ChildMaxScore(ref node.Children); sim.CompletionState,
sim.AvailableActionsHeuristic(strict)
if (node.State.Action != null) ));
actions.Add(node.State.Action.Value);
}
return (actions, node.State);
} }
public static (List<ActionType> Actions, SimulationState State) SearchStepwise(SolverConfig config, SimulationInput input, Action<ActionType>? actionCallback, CancellationToken token = default) => public static (List<ActionType> Actions, SimulationState State) SearchStepwise<S>(SolverConfig config, SimulationInput input, Action<ActionType>? actionCallback, CancellationToken token = default) where S : ISolver =>
SearchStepwise(config, new SimulationState(input), actionCallback, token); SearchStepwise<S>(config, new SimulationState(input), actionCallback, token);
public static (List<ActionType> Actions, SimulationState State) SearchStepwise(SolverConfig config, SimulationState state, Action<ActionType>? actionCallback, CancellationToken token = default) public static (List<ActionType> Actions, SimulationState State) SearchStepwise<S>(SolverConfig config, SimulationState state, Action<ActionType>? actionCallback, CancellationToken token = default) where S : ISolver
{ {
var actions = new List<ActionType>(); var actions = new List<ActionType>();
Simulator sim = new(state, config.MaxStepCount); var sim = new Simulator(state, config.MaxStepCount);
var solver = new Solver(config, state, true); var rootNode = CreateRootNode(config, state, true);
while (!sim.IsComplete) while (!sim.IsComplete)
{ {
if (token.IsCancellationRequested) if (token.IsCancellationRequested)
break; break;
solver.Search(token); S.Search(ref config, rootNode, token);
var (solution_actions, solution_node) = solver.Solution(); var (solution_actions, solution_node) = Solution(rootNode);
if (solution_node.Scores.MaxScore >= 1.0) if (solution_node.Scores.MaxScore >= 1.0)
{ {
@@ -298,20 +231,20 @@ public sealed class Solver
actionCallback?.Invoke(chosen_action); actionCallback?.Invoke(chosen_action);
solver = new Solver(config, state, true); rootNode = CreateRootNode(config, state, true);
} }
return (actions, state); return (actions, state);
} }
public static (List<ActionType> Actions, SimulationState State) SearchOneshot(SolverConfig config, SimulationInput input, CancellationToken token = default) => public static (List<ActionType> Actions, SimulationState State) SearchOneshot<S>(SolverConfig config, SimulationInput input, CancellationToken token = default) where S : ISolver =>
SearchOneshot(config, new SimulationState(input), token); SearchOneshot<S>(config, new SimulationState(input), token);
public static (List<ActionType> Actions, SimulationState State) SearchOneshot(SolverConfig config, SimulationState state, CancellationToken token = default) public static (List<ActionType> Actions, SimulationState State) SearchOneshot<S>(SolverConfig config, SimulationState state, CancellationToken token = default) where S : ISolver
{ {
var solver = new Solver(config, state, false); var rootNode = CreateRootNode(config, state, false);
solver.Search(token); S.Search(ref config, rootNode, token);
var (solution_actions, solution_node) = solver.Solution(); var (solution_actions, solution_node) = Solution(rootNode);
return (solution_actions, solution_node.State); return (solution_actions, solution_node.State);
} }
} }