beet_ml 0.0.8

Machine Learning actions built upon beet_flow
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129

use std::marker::PhantomData;

use crate::prelude::*;
use beet_core::prelude::*;
use beet_flow::prelude::*;
use beet_spatial::prelude::*;

/// Finds the [`Sentence`] with the highest similarity to the agent's,
/// then set it as the agent's [`SteerTarget`].
/// The generic parameter is used to [`With`] filter the entities to consider.
#[action(sentence_steer_target::<F>)]
#[derive(Debug, Clone, PartialEq, Component, Reflect)]
#[reflect(Default, Component)]
// TODO OnRun<Sentence>
// RunResult<SteerTarget?>

pub struct SentenceSteerTarget<F: Component> {
	pub target_entity: TargetEntity,
	// / The value below which the agent will ignore the target.
	// pub threshold:f32,
	// #[serde(bound = "")]
	_phantom: PhantomData<F>,
}

impl<F: Component> SentenceSteerTarget<F> {
	/// Create a new [`SentenceSteerTarget`] with the given [`TargetEntity`].
	pub fn new(target_entity: TargetEntity) -> Self {
		Self {
			target_entity,
			_phantom: PhantomData,
		}
	}
}

impl<F: Component> Default for SentenceSteerTarget<F> {
	fn default() -> Self {
		Self {
			target_entity: Default::default(),
			_phantom: PhantomData,
		}
	}
}

fn sentence_steer_target<F: Component>(
	ev: On<GetOutcome>,
	mut commands: Commands,
	query: Query<(&HandleWrapper<Bert>, &SentenceSteerTarget<F>)>,
	sentences: Query<&Sentence>,
	// TODO this should be query of Sentence, but we need
	// it to be similar to sentence_scorer
	items: Query<Entity, (With<Sentence>, With<F>)>,
	mut berts: ResMut<Assets<Bert>>,
) -> Result {
	let (handle, sentence_steer_target) = query.get(ev.action())?;

	let target_entity = sentence_steer_target.target_entity.select_target(&ev);

	let target_sentence = sentences.get(target_entity)?;

	let bert = berts
		.get_mut(handle)
		.expect(&expect_action::to_have_asset(&ev));

	match bert.closest_sentence_entity(
		target_sentence.0.clone(),
		items
			.into_iter()
			.filter(|e| *e != target_entity)
			.collect::<Vec<_>>(),
		&sentences,
	) {
		Ok(entity) => {
			commands
				.entity(ev.agent())
				.insert(SteerTarget::Entity(entity));
		}
		Err(e) => log::error!("SentenceFlow: {}", e),
	}
	Ok(())
}

// #[cfg(test)]
// mod test {
// 	use crate::prelude::*;
// 	use beet_flow::prelude::*;
// 	use beet_spatial::steer::SteerTarget;
// 	use beet_core::prelude::*;
// 	use beet_core::prelude::*;

// 	#[test]
// 	fn works() {
// 		pretty_env_logger::try_init().ok();

// 		let mut app = App::new();
// 		app.add_plugins((
// 			MinimalPlugins,
// 			ControlFlowPlugin::default(),
// 			workspace_asset_plugin(),
// 			Language::default(),
// 		))
// 		.finish();

// 		let handle =
// 			block_on_asset_load::<Bert>(&mut app, "ml/default-bert.ron")
// 				.unwrap();

// 		let world = app.world_mut();

// 		let agent = world.spawn_empty().id();

// 		let heal = world.spawn(Sentence::new("heal")).id();
// 		let kill = world.spawn(Sentence::new("kill")).id();

// 		let behavior = world
// 			.spawn((
// 				TargetEntity(agent),
// 				InsertSentenceSteerTarget::<Sentence>::default(),
// 				HandleWrapper(handle),
// 			))
// 			.id();
// 		world.flush();
// 		world.entity_mut(behavior).insert(Sentence::new("destroy"));
// 		world.flush();

// 		let target = world.entity(agent).get::<SteerTarget>();
// 		target.xpect_not_eq(Some(&SteerTarget::Entity(heal)));
// 		target.xpect_eq(Some(&SteerTarget::Entity(kill)));
// 	}
// }