File size: 8,218 Bytes
826dd96 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 | from __future__ import annotations
from pathlib import Path
from autofarm.contracts import (
ChallengeState,
EpisodeOutcome,
EpisodeOutcomeStatus,
FieldZoneState,
PriorityBand,
RiddleCategory,
SimulatorMode,
)
from autofarm.sim.engine import (
DEFAULT_HOME_ZONE_ID,
DEFAULT_INTERACTIVE_BASE_STATION_ZONE_ID,
DEFAULT_INTERACTIVE_MAX_ENERGY,
HiddenZoneState,
SimulatorEnvironment,
clone_hidden_zone_state,
default_hidden_for_zone,
)
from autofarm.sim.scenarios import ChallengeTemplate, InspectFixture, RIDDLE_SAFETY_MAX_STEPS, RiddleSpec, ZoneScenarioTruth
from autofarm.sim.scenarios import available_challenge_templates, fallback_goal_policy
class EmptyCatalog:
def sample_by_identity(self, *, dataset_name: str, source_dataset: str, sample_id: str): # noqa: ARG002
return None
def samples_for_tag(self, dataset_name: str, tag: str, positive: bool = True): # noqa: ARG002
return []
def samples_for_dataset(self, dataset_name: str): # noqa: ARG002
return []
def make_fixture(*expected_anomalies: str, simulated_backend_error: str | None = None) -> InspectFixture:
return InspectFixture(
sample_ref=None,
expected_anomalies=tuple(expected_anomalies),
simulated_backend_error=simulated_backend_error,
)
def make_environment(
*,
name: str,
zone_id: str = "zone_r02_c02",
priority_band: PriorityBand = PriorityBand.HIGH,
priority_hint: float = 0.88,
state_confidence: float = 0.4,
starting_uncertainty: float = 0.6,
fixtures: tuple[InspectFixture, ...] = (),
travel_budget: float = 10.0,
) -> SimulatorEnvironment:
template_policy = available_challenge_templates().get(name)
goal_policy = (
template_policy.goal_policy
if template_policy is not None
else fallback_goal_policy().model_copy(
update={
"allowed_terminal_intents": [],
"required_positive_confirmation_count": 999,
}
)
)
zone_states = build_test_zone_grid(focus_zone_id=zone_id)
zone_state = next(zone for zone in zone_states if zone.zone_id == zone_id)
zone_state.static_features["scenario_priority_hint"] = priority_hint
zone_state.state_confidence = state_confidence
truth = ZoneScenarioTruth(
true_priority_band=priority_band,
inspect_fixtures=fixtures,
priority_hint=priority_hint,
confidence_override=state_confidence,
starting_uncertainty=starting_uncertainty,
)
template = ChallengeTemplate(
template_id=name,
display_name=name,
description=name,
resolver_id=name,
truth=truth,
goal_policy=goal_policy,
interactive_enabled=True,
)
riddle = RiddleSpec(
name=name,
description=name,
category=RiddleCategory.MISSION,
travel_budget=travel_budget,
challenge_template_id=name,
target_zone_id=zone_id,
evaluator_id=name,
objective=name,
challenge_summary=name,
why_it_matters=name,
)
hidden = HiddenZoneState(
zone_id=zone_id,
true_priority_band=priority_band,
inspect_fixtures=fixtures,
current_uncertainty=starting_uncertainty,
)
default_hidden_zone_states = {state.zone_id: default_hidden_for_zone(state) for state in zone_states}
hidden_zone_states = {
state.zone_id: default_hidden_for_zone(state)
for state in zone_states
}
hidden_zone_states[zone_id] = hidden
baseline_zone_states = {state.zone_id: state.model_copy(deep=True) for state in zone_states}
env = SimulatorEnvironment(
mode=SimulatorMode.RIDDLE,
name=name,
description=name,
challenge_templates={name: template},
zone_states=zone_states,
hidden_zone_states=hidden_zone_states,
baseline_zone_states=baseline_zone_states,
default_hidden_zone_states=default_hidden_zone_states,
catalog=EmptyCatalog(),
current_zone_id=zone_id,
travel_budget_remaining=travel_budget,
max_energy=DEFAULT_INTERACTIVE_MAX_ENERGY,
energy_remaining=DEFAULT_INTERACTIVE_MAX_ENERGY,
riddle_spec=riddle,
outcome=EpisodeOutcome(
status=EpisodeOutcomeStatus.IN_PROGRESS,
done_reason="running",
message="Riddle in progress.",
safety_step_cap=RIDDLE_SAFETY_MAX_STEPS,
),
active_challenges={
zone_id: ChallengeState(
instance_id=f"{name}:{zone_id}:test",
template_id=name,
name=name,
description=name,
zone_id=zone_id,
evaluator_id=name,
goal_policy=goal_policy,
)
},
last_visited_step_by_zone={state.zone_id: (0 if state.zone_id == zone_id else -1) for state in zone_states},
)
env.initialize_indexes()
return env
def make_interactive_environment(
*,
zone_id: str = DEFAULT_INTERACTIVE_BASE_STATION_ZONE_ID,
max_energy: float = DEFAULT_INTERACTIVE_MAX_ENERGY,
travel_budget: float | None = None,
) -> SimulatorEnvironment:
del max_energy
zone_states = build_test_zone_grid(focus_zone_id=zone_id)
default_hidden_zone_states = {state.zone_id: default_hidden_for_zone(state) for state in zone_states}
env = SimulatorEnvironment(
mode=SimulatorMode.INTERACTIVE,
name="interactive_world",
description="interactive_world",
challenge_templates={},
zone_states=zone_states,
hidden_zone_states={zone_id: clone_hidden_zone_state(hidden) for zone_id, hidden in default_hidden_zone_states.items()},
baseline_zone_states={state.zone_id: state.model_copy(deep=True) for state in zone_states},
default_hidden_zone_states=default_hidden_zone_states,
catalog=EmptyCatalog(),
current_zone_id=DEFAULT_HOME_ZONE_ID,
travel_budget_remaining=travel_budget,
home_zone_id=DEFAULT_HOME_ZONE_ID,
home_neighbor_zone_id=zone_id,
max_energy=DEFAULT_INTERACTIVE_MAX_ENERGY,
energy_remaining=DEFAULT_INTERACTIVE_MAX_ENERGY,
outcome=EpisodeOutcome(
status=EpisodeOutcomeStatus.IN_PROGRESS,
done_reason="running",
message="Interactive world is running.",
),
last_visited_step_by_zone={state.zone_id: (0 if state.zone_id == zone_id else -1) for state in zone_states},
)
env.initialize_indexes()
return env
def parse_zone_coords(zone_id: str) -> tuple[int, int]:
row_token = zone_id.split("_")[1]
col_token = zone_id.split("_")[2]
return int(row_token.replace("r", "")), int(col_token.replace("c", ""))
def build_test_zone_grid(*, focus_zone_id: str) -> list[FieldZoneState]:
zone_states: list[FieldZoneState] = []
for row in range(1, 4):
for col in range(1, 4):
zone_id = f"zone_r{row:02d}_c{col:02d}"
zone_states.append(
FieldZoneState(
zone_id=zone_id,
timestamp="2026-04-01",
static_features={
"scenario_priority_hint": 0.25 if zone_id != focus_zone_id else 0.4,
"recommended_action": "none",
},
weather_proxies={},
soil_source_status="usda_sda_exact_point",
state_confidence=0.75,
recent_findings=[],
row=row,
col=col,
)
)
by_coord = {(zone.row, zone.col): zone.zone_id for zone in zone_states}
enriched: list[FieldZoneState] = []
for zone in zone_states:
neighbors: list[str] = []
for row_delta, col_delta in ((-1, 0), (0, 1), (1, 0), (0, -1)):
neighbor_zone_id = by_coord.get((zone.row + row_delta, zone.col + col_delta))
if neighbor_zone_id is not None:
neighbors.append(neighbor_zone_id)
enriched.append(zone.model_copy(update={"neighbor_zone_ids": neighbors}))
return enriched
|