File size: 13,629 Bytes
fcf8749 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 | """
Unit tests for Explainability Agent v2 (Phase 4.3).
Tests classification logic and template generation.
"""
import pytest
from app.services.explainability import ExplainabilityAgent
from app.schemas.explainability import DriverExplanationInput, DriverExplanationOutput
class TestCategoryClassification:
"""Tests for _classify_category logic."""
def setup_method(self):
"""Set up test fixtures."""
self.agent = ExplainabilityAgent()
def _base_input(self, **overrides) -> DriverExplanationInput:
"""Create base input with optional overrides."""
defaults = {
"driver_id": "driver-1",
"driver_name": "Test Driver",
"num_drivers": 5,
"today_effort": 50.0,
"today_rank": 3,
"route_id": "route-1",
"route_summary": {
"num_packages": 20,
"total_weight_kg": 50.0,
"num_stops": 10,
"difficulty_score": 2.5,
"estimated_time_minutes": 180,
},
"effort_breakdown": {
"physical_effort": 15.0,
"route_complexity": 10.0,
"time_pressure": 5.0,
},
"global_avg_effort": 50.0,
"global_std_effort": 10.0,
"global_gini_index": 0.15,
"global_max_gap": 20.0,
"history_efforts_last_7_days": [45.0, 50.0, 55.0],
"history_hard_days_last_7": 1,
"is_recovery_day": False,
"had_manual_override": False,
"liaison_decision": None,
"swap_applied": False,
}
defaults.update(overrides)
return DriverExplanationInput(**defaults)
def test_near_avg_category(self):
"""Effort close to average should classify as NEAR_AVG."""
data = self._base_input(today_effort=52.0, global_avg_effort=50.0)
category = self.agent._classify_category(data)
assert category == "NEAR_AVG"
def test_heavy_category(self):
"""Above-average effort without negotiation should classify as HEAVY."""
data = self._base_input(today_effort=70.0, global_avg_effort=50.0)
category = self.agent._classify_category(data)
assert category == "HEAVY"
def test_heavy_with_swap_category(self):
"""Above-average with swap applied should classify as HEAVY_WITH_SWAP."""
data = self._base_input(
today_effort=70.0,
global_avg_effort=50.0,
swap_applied=True,
)
category = self.agent._classify_category(data)
assert category == "HEAVY_WITH_SWAP"
def test_heavy_no_swap_category(self):
"""Above-average with COUNTER but no swap should classify as HEAVY_NO_SWAP."""
data = self._base_input(
today_effort=70.0,
global_avg_effort=50.0,
liaison_decision="COUNTER",
swap_applied=False,
)
category = self.agent._classify_category(data)
assert category == "HEAVY_NO_SWAP"
def test_recovery_category(self):
"""Explicit recovery day should classify as RECOVERY."""
data = self._base_input(
today_effort=35.0,
global_avg_effort=50.0,
is_recovery_day=True,
)
category = self.agent._classify_category(data)
assert category == "RECOVERY"
def test_light_recovery_category(self):
"""Below average with hard day streak should classify as LIGHT_RECOVERY."""
data = self._base_input(
today_effort=35.0,
global_avg_effort=50.0,
history_hard_days_last_7=3,
)
category = self.agent._classify_category(data)
assert category == "LIGHT_RECOVERY"
def test_light_category(self):
"""Below average without hard streak should classify as LIGHT."""
data = self._base_input(
today_effort=35.0,
global_avg_effort=50.0,
history_hard_days_last_7=0,
)
category = self.agent._classify_category(data)
assert category == "LIGHT"
class TestDriverTextGeneration:
"""Tests for _build_driver_text templates."""
def setup_method(self):
"""Set up test fixtures."""
self.agent = ExplainabilityAgent()
def _base_input(self, **overrides) -> DriverExplanationInput:
"""Create base input with optional overrides."""
defaults = {
"driver_id": "driver-1",
"driver_name": "Test Driver",
"num_drivers": 5,
"today_effort": 50.0,
"today_rank": 3,
"route_id": "route-1",
"route_summary": {
"num_packages": 20,
"total_weight_kg": 50.0,
"num_stops": 10,
"difficulty_score": 2.5,
"estimated_time_minutes": 180,
},
"effort_breakdown": {},
"global_avg_effort": 50.0,
"global_std_effort": 10.0,
"global_gini_index": 0.15,
"global_max_gap": 20.0,
"history_efforts_last_7_days": [],
"history_hard_days_last_7": 0,
"is_recovery_day": False,
"had_manual_override": False,
"liaison_decision": None,
"swap_applied": False,
}
defaults.update(overrides)
return DriverExplanationInput(**defaults)
def test_near_avg_contains_expected_phrases(self):
"""NEAR_AVG text should mention 'moderate' and 'balanced'."""
data = self._base_input()
text = self.agent._build_driver_text(data, "NEAR_AVG")
assert "moderate" in text.lower()
assert "balanced" in text.lower()
assert "20 packages" in text
assert "10 stops" in text
def test_heavy_mentions_heavier(self):
"""HEAVY category text should mention 'heavier'."""
data = self._base_input(today_effort=70.0)
text = self.agent._build_driver_text(data, "HEAVY")
assert "heavier" in text.lower()
def test_recovery_mentions_lighter(self):
"""RECOVERY text should mention 'intentionally lighter' and 'recover'."""
data = self._base_input(is_recovery_day=True)
text = self.agent._build_driver_text(data, "RECOVERY")
assert "lighter" in text.lower()
assert "recover" in text.lower()
def test_includes_packages_and_stops(self):
"""All templates should include package and stop counts."""
data = self._base_input()
for category in ["NEAR_AVG", "HEAVY", "LIGHT", "RECOVERY"]:
text = self.agent._build_driver_text(data, category)
assert "20 packages" in text
assert "10 stops" in text
class TestAdminTextGeneration:
"""Tests for _build_admin_text templates."""
def setup_method(self):
"""Set up test fixtures."""
self.agent = ExplainabilityAgent()
def _base_input(self, **overrides) -> DriverExplanationInput:
"""Create base input with optional overrides."""
defaults = {
"driver_id": "driver-1",
"driver_name": "John Doe",
"num_drivers": 5,
"today_effort": 65.0,
"today_rank": 2,
"route_id": "route-1",
"route_summary": {
"num_packages": 25,
"total_weight_kg": 75.0,
"num_stops": 12,
"difficulty_score": 3.0,
"estimated_time_minutes": 240,
},
"effort_breakdown": {
"physical_effort": 25.0,
"route_complexity": 20.0,
"time_pressure": 10.0,
},
"global_avg_effort": 50.0,
"global_std_effort": 12.0,
"global_gini_index": 0.12,
"global_max_gap": 25.0,
"history_efforts_last_7_days": [55.0, 60.0],
"history_hard_days_last_7": 2,
"is_recovery_day": False,
"had_manual_override": False,
"liaison_decision": None,
"swap_applied": False,
}
defaults.update(overrides)
return DriverExplanationInput(**defaults)
def test_admin_text_includes_driver_name(self):
"""Admin text should include driver name."""
data = self._base_input()
text = self.agent._build_admin_text(data, "NEAR_AVG")
assert "John Doe" in text
def test_admin_text_includes_effort_metrics(self):
"""Admin text should include effort score and comparison to average."""
data = self._base_input()
text = self.agent._build_admin_text(data, "NEAR_AVG")
assert "65" in text # effort score
assert "50" in text # average
def test_admin_text_includes_gini(self):
"""Admin text should include Gini index."""
data = self._base_input()
text = self.agent._build_admin_text(data, "NEAR_AVG")
assert "Gini" in text
assert "0.12" in text
def test_admin_text_includes_rank(self):
"""Admin text should include rank."""
data = self._base_input()
text = self.agent._build_admin_text(data, "NEAR_AVG")
assert "2/5" in text # rank/total
def test_heavy_with_swap_mentions_swap(self):
"""HEAVY_WITH_SWAP should mention swap in admin text."""
data = self._base_input(swap_applied=True)
text = self.agent._build_admin_text(data, "HEAVY_WITH_SWAP")
assert "swap" in text.lower()
def test_recovery_mentions_hard_days(self):
"""RECOVERY text should mention hard days count."""
data = self._base_input(is_recovery_day=True, history_hard_days_last_7=4)
text = self.agent._build_admin_text(data, "RECOVERY")
assert "4" in text
assert "hard day" in text.lower() or "recovery" in text.lower()
def test_manual_override_note_included(self):
"""Manual override should be noted in admin text."""
data = self._base_input(had_manual_override=True)
text = self.agent._build_admin_text(data, "NEAR_AVG")
assert "override" in text.lower()
class TestBuildExplanationForDriver:
"""Tests for the main build_explanation_for_driver method."""
def setup_method(self):
"""Set up test fixtures."""
self.agent = ExplainabilityAgent()
def test_returns_correct_output_type(self):
"""Should return DriverExplanationOutput."""
data = DriverExplanationInput(
driver_id="d1",
driver_name="Driver One",
num_drivers=3,
today_effort=50.0,
today_rank=2,
route_id="r1",
route_summary={"num_packages": 15, "total_weight_kg": 40.0, "num_stops": 8, "difficulty_score": 2.0, "estimated_time_minutes": 120},
global_avg_effort=50.0,
global_std_effort=10.0,
global_gini_index=0.1,
global_max_gap=15.0,
)
result = self.agent.build_explanation_for_driver(data)
assert isinstance(result, DriverExplanationOutput)
assert isinstance(result.driver_explanation, str)
assert isinstance(result.admin_explanation, str)
assert isinstance(result.category, str)
def test_driver_and_admin_texts_are_different(self):
"""Driver and admin explanations should have different detail levels."""
data = DriverExplanationInput(
driver_id="d1",
driver_name="Driver One",
num_drivers=3,
today_effort=50.0,
today_rank=2,
route_id="r1",
route_summary={"num_packages": 15, "total_weight_kg": 40.0, "num_stops": 8, "difficulty_score": 2.0, "estimated_time_minutes": 120},
global_avg_effort=50.0,
global_std_effort=10.0,
global_gini_index=0.1,
global_max_gap=15.0,
)
result = self.agent.build_explanation_for_driver(data)
# Admin text should be longer and contain more metrics
assert len(result.admin_explanation) > len(result.driver_explanation)
assert "Gini" in result.admin_explanation
assert "Gini" not in result.driver_explanation
class TestSnapshotGeneration:
"""Tests for DecisionLog snapshot helpers."""
def setup_method(self):
"""Set up test fixtures."""
self.agent = ExplainabilityAgent()
def test_input_snapshot(self):
"""Input snapshot should include key metrics."""
snapshot = self.agent.get_input_snapshot(
num_drivers=5,
avg_effort=55.0,
std_effort=12.0,
gini_index=0.15,
category_counts={"NEAR_AVG": 3, "HEAVY": 2},
)
assert snapshot["num_drivers"] == 5
assert snapshot["avg_effort"] == 55.0
assert snapshot["gini_index"] == 0.15
def test_output_snapshot(self):
"""Output snapshot should include totals and category counts."""
category_counts = {"NEAR_AVG": 3, "HEAVY": 2}
snapshot = self.agent.get_output_snapshot(
total_explanations=5,
category_counts=category_counts,
)
assert snapshot["total_explanations"] == 5
assert snapshot["category_counts"]["NEAR_AVG"] == 3
assert snapshot["category_counts"]["HEAVY"] == 2
|