Day 73 (12:00): auto-format backend

Author: Test User

src/backend/api/chat.py

@@ -2406,9 +2406,7 @@ def _detect_interaction_request(
 )
 
 _GOAL_SEEK_NUMBER_RE = re.compile(
-    r"(?:^|[\s=:to])"
-    r"(?:[$€£¥]?\s*)?"
-    r"(\d[\d,]*(?:\.\d+)?)\s*([kmb])?\b",
+    r"(?:^|[\s=:to])" r"(?:[$€£¥]?\s*)?" r"(\d[\d,]*(?:\.\d+)?)\s*([kmb])?\b",
     re.IGNORECASE,
 )
 
@@ -11968,9 +11966,7 @@ def _fmt_metric(metrics_json: str | None) -> str | None:
             _wu_trend = (
                 "up"
                 if (_wu_change_pct or 0) > 5
-                else "down"
-                if (_wu_change_pct or 0) < -5
-                else "flat"
+                else "down" if (_wu_change_pct or 0) < -5 else "flat"
             )
 
             # Per-day breakdown for the current week (7 entries)
@@ -11989,7 +11985,9 @@ def _fmt_metric(metrics_json: str | None) -> str | None:
             _wu_feature_tally: dict[str, dict[str, int]] = {}
             _wu_recent_logs = [
                 lg for lg in _wu_logs if lg.created_at >= _wu_week_start
-            ][:100]  # cap to last 100 for performance
+            ][
+                :100
+            ]  # cap to last 100 for performance
             for _wl in _wu_recent_logs:
                 try:
                     _feat_dict: dict = json.loads(_wl.input_features or "{}")
@@ -14627,9 +14625,7 @@ def _fmt_metric(metrics_json: str | None) -> str | None:
                         else (
                             "healthy"
                             if _n_failed == 0
-                            else "warning"
-                            if _n_failed / _n_total < 0.1
-                            else "critical"
+                            else "warning" if _n_failed / _n_total < 0.1 else "critical"
                         )
                     )
                     _wh_total_events += _n_total
@@ -14688,9 +14684,7 @@ def _fmt_metric(metrics_json: str | None) -> str | None:
                     else (
                         "warning"
                         if any(d["status"] == "warning" for d in _wh_dep_summaries)
-                        else "no_events"
-                        if _wh_total_events == 0
-                        else "healthy"
+                        else "no_events" if _wh_total_events == 0 else "healthy"
                     )
                 )
             )

src/backend/api/deploy.py

@@ -2714,9 +2714,7 @@ def set_accuracy_alert(
             + (
                 f"drops below {thr:.0%}."
                 if problem_type == "classification" and thr is not None
-                else f"exceeds {thr:.1f}%."
-                if thr is not None
-                else ""
+                else f"exceeds {thr:.1f}%." if thr is not None else ""
             )
             if threshold_set
             else "Accuracy alert disabled."

src/backend/api/projects.py

@@ -158,9 +158,7 @@ def get_portfolio(session: Session = Depends(get_session)):
             metric_name = (
                 "r2"
                 if "r2" in metrics
-                else "accuracy"
-                if "accuracy" in metrics
-                else None
+                else "accuracy" if "accuracy" in metrics else None
             )
             if metric_val is not None and (
                 best_metric_value is None or metric_val > best_metric_value
@@ -241,9 +239,7 @@ def get_cross_project_comparison(session: Session = Depends(get_session)):
             metric_name = (
                 "r2"
                 if "r2" in metrics
-                else "accuracy"
-                if "accuracy" in metrics
-                else None
+                else "accuracy" if "accuracy" in metrics else None
             )
             if metric_val is not None and (
                 best_metric_value is None or metric_val > best_metric_value
@@ -591,9 +587,7 @@ def _static_narrative(ctx: dict) -> str:
         quality = (
             "clean"
             if d["missing_pct"] < 5
-            else "mostly complete"
-            if d["missing_pct"] < 20
-            else "with some gaps"
+            else "mostly complete" if d["missing_pct"] < 20 else "with some gaps"
         )
         parts.append(
             f"The analysis is based on **{d['filename']}** — "

src/backend/core/advisor.py

@@ -364,23 +364,15 @@ def _build_summary(
         quality = (
             "excellent"
             if pct >= 85
-            else "good"
-            if pct >= 70
-            else "moderate"
-            if pct >= 55
-            else "limited"
+            else "good" if pct >= 70 else "moderate" if pct >= 55 else "limited"
         )
         base = f"Your model explains {pct}% of variation in the target — {quality} predictive power."
     else:
         pct = round(primary_metric * 100)
         quality = (
             "excellent"
             if pct >= 90
-            else "good"
-            if pct >= 80
-            else "moderate"
-            if pct >= 70
-            else "limited"
+            else "good" if pct >= 80 else "moderate" if pct >= 70 else "limited"
         )
         base = f"Your model achieves {pct}% accuracy — {quality} performance."
 

src/backend/core/deployer.py

@@ -1043,9 +1043,7 @@ def compute_prediction_cohort(
         direction_label = (
             "higher"
             if top_mean > overall_mean
-            else "lower"
-            if top_mean < overall_mean
-            else "similar"
+            else "lower" if top_mean < overall_mean else "similar"
         )
         numeric_profile.append(
             {

src/backend/core/query_engine.py

@@ -285,9 +285,7 @@ def _execute_spec(spec: dict, df: pd.DataFrame) -> tuple[pd.DataFrame | None, st
         strength = (
             "strong"
             if abs(corr_val) >= 0.7
-            else "moderate"
-            if abs(corr_val) >= 0.4
-            else "weak"
+            else "moderate" if abs(corr_val) >= 0.4 else "weak"
         )
         direction = "positive" if corr_val >= 0 else "negative"
         text = (

src/backend/core/report_generator.py

@@ -403,9 +403,7 @@ def generate_model_card_html(
         score_color = (
             "#059669"
             if brier_score < 0.1
-            else "#d97706"
-            if brier_score < 0.2
-            else "#dc2626"
+            else "#d97706" if brier_score < 0.2 else "#dc2626"
         )
         cal_html = f"""<h2 style="font-size:1rem;font-weight:600;color:#1e40af;margin:1.5rem 0 0.5rem">
             Calibration

src/backend/tests/test_accuracy_alert.py

@@ -162,9 +162,9 @@ def test_accuracy_alert_patterns_no_false_positives():
         "what is the drift score",
     ]
     for phrase in non_matches:
-        assert not _ACCURACY_ALERT_PATTERNS.search(phrase), (
-            f"False positive: {phrase!r}"
-        )
+        assert not _ACCURACY_ALERT_PATTERNS.search(
+            phrase
+        ), f"False positive: {phrase!r}"
 
 
 def test_accuracy_alert_threshold_re_extracts_percent():

src/backend/tests/test_anomaly.py

@@ -246,6 +246,6 @@ def test_irrelevant_message_does_not_match(self):
             "deploy the model",
         ]
         for phrase in phrases:
-            assert not _ANOMALY_PATTERNS.search(phrase), (
-                f"Unexpected match for: {phrase!r}"
-            )
+            assert not _ANOMALY_PATTERNS.search(
+                phrase
+            ), f"Unexpected match for: {phrase!r}"

src/backend/tests/test_balance_train_chat.py

@@ -219,9 +219,9 @@ def test_class_weight_emits_training_started(self, sync_client, tmp_path):
         pid = _setup_classification(sync_client, tmp_path)
         events = _chat_events(sync_client, pid, "train with class weighting")
         started = [e for e in events if e.get("type") == "training_started"]
-        assert len(started) == 1, (
-            f"Expected training_started, got: {[e.get('type') for e in events]}"
-        )
+        assert (
+            len(started) == 1
+        ), f"Expected training_started, got: {[e.get('type') for e in events]}"
         payload = started[0]["training"]
         assert payload["imbalance_strategy"] == "class_weight"
         assert payload["problem_type"] == "classification"
@@ -256,9 +256,9 @@ def test_regression_target_no_training_started(self, sync_client, tmp_path):
         pid = _setup_regression(sync_client, tmp_path)
         events = _chat_events(sync_client, pid, "train with class weighting")
         started = [e for e in events if e.get("type") == "training_started"]
-        assert len(started) == 0, (
-            "Should not start training for regression with imbalance correction"
-        )
+        assert (
+            len(started) == 0
+        ), "Should not start training for regression with imbalance correction"
 
     def test_fix_imbalance_variant_works(self, sync_client, tmp_path):
         """'fix the imbalance and train' phrase triggers balanced training."""