Phase 2: Complete output standardization

- Added suffix to diagnostic columns
- Fixed 4 tests for new column names
- 40 tests passing
- Merge compatibility achieved

Author: miranov25

UTILS/dfextensions/groupby_regression/groupby_regression_optimized.py

@@ -83,11 +83,11 @@ def _make_nan_result_row(
 
     # Add diagnostics (if dict provided)
     if diag_dict:
-        row[f"{diag_prefix}n_total"] = diag_dict.get('n_total', 0)
-        row[f"{diag_prefix}n_valid"] = diag_dict.get('n_valid', 0)
-        row[f"{diag_prefix}n_filtered"] = diag_dict.get('n_filtered', 0)
-        row[f"{diag_prefix}cond_xtx"] = diag_dict.get('cond_xtx', np.inf)
-        row[f"{diag_prefix}status"] = diag_dict.get('status', 'UNKNOWN')
+        row[f"{diag_prefix}n_total{suffix}"] = diag_dict.get('n_total', 0)
+        row[f"{diag_prefix}n_valid{suffix}"] = diag_dict.get('n_valid', 0)
+        row[f"{diag_prefix}n_filtered{suffix}"] = diag_dict.get('n_filtered', 0)
+        row[f"{diag_prefix}cond_xtx{suffix}"] = diag_dict.get('cond_xtx', np.inf)
+        row[f"{diag_prefix}status{suffix}"] = diag_dict.get('status', 'UNKNOWN')
 
     return row
 
@@ -975,12 +975,12 @@ def make_parallel_fit_v3(
 
         # Add diagnostics (if enabled)
         if diag:
-            row[f"{diag_prefix}n_total"] = n_total
-            row[f"{diag_prefix}n_valid"] = n_valid
-            row[f"{diag_prefix}n_filtered"] = n_filtered
-            row[f"{diag_prefix}cond_xtx"] = cond
-            row[f"{diag_prefix}status"] = status
-            row[f"{diag_prefix}time_ms"] = (t1 - t0) * 1000
+            row[f"{diag_prefix}n_total{suffix}"] = n_total
+            row[f"{diag_prefix}n_valid{suffix}"] = n_valid
+            row[f"{diag_prefix}n_filtered{suffix}"] = n_filtered
+            row[f"{diag_prefix}cond_xtx{suffix}"] = cond
+            row[f"{diag_prefix}status{suffix}"] = status
+            row[f"{diag_prefix}time_ms{suffix}"] = (t1 - t0) * 1000
 
         res_rows.append(row)
 
@@ -993,7 +993,7 @@ def make_parallel_fit_v3(
     # Add wall time diagnostic
     if diag:
         t_end = time.perf_counter()
-        dfGB[f"{diag_prefix}wall_ms"] = (t_end - t_start) * 1000
+        dfGB[f"{diag_prefix}wall_ms{suffix}"] = (t_end - t_start) * 1000
 
     # ========================================================================
     # 5. HANDLE PREDICTIONS (IF REQUESTED)
@@ -1357,11 +1357,11 @@ def make_parallel_fit_v4(
             out_dict[f"{tname}_mad{suffix}"] = mad_arr[:, t_idx]
     # Diagnostics (if enabled)
     if diag:
-        out_dict[f"{diag_prefix}n_total"] = n_total_arr
-        out_dict[f"{diag_prefix}n_valid"] = n_valid_arr
-        out_dict[f"{diag_prefix}n_filtered"] = n_filtered_arr
-        out_dict[f"{diag_prefix}cond_xtx"] = cond_arr
-        out_dict[f"{diag_prefix}status"] = status_arr
+        out_dict[f"{diag_prefix}n_total{suffix}"] = n_total_arr
+        out_dict[f"{diag_prefix}n_valid{suffix}"] = n_valid_arr
+        out_dict[f"{diag_prefix}n_filtered{suffix}"] = n_filtered_arr
+        out_dict[f"{diag_prefix}cond_xtx{suffix}"] = cond_arr
+        out_dict[f"{diag_prefix}status{suffix}"] = status_arr
 
 
 

UTILS/dfextensions/groupby_regression/tests/test_groupby_regression_optimized.py

@@ -1597,14 +1597,14 @@ def test_v3_inf_nan_filtering():
     )
 
     # Check diagnostics
-    assert 'diag_n_total' in dfGB.columns
-    assert 'diag_n_filtered' in dfGB.columns
+    assert 'diag_n_total_test' in dfGB.columns
+    assert 'diag_n_filtered_test' in dfGB.columns
 
     # Each group should have filtered 1 row
-    assert all(dfGB['diag_n_filtered'] == 1)
+    assert all(dfGB['diag_n_filtered_test'] == 1)
 
     # All should have status OK (enough data remains)
-    assert all(dfGB['diag_status'] == 'OK')
+    assert all(dfGB['diag_status_test'] == 'OK')
 
     print("✅ test_v3_inf_nan_filtering passed")
 
@@ -1820,12 +1820,12 @@ def test_v3_singular_matrix_handling():
     )
 
     # Check status indicates problem
-    status = dfGB['diag_status'].values[0]
+    status = dfGB['diag_status_test'].values[0]
     assert status in ['ILL_CONDITIONED_RIDGED', 'SINGULAR_MATRIX'], \
         f"Expected ill-conditioned status, got {status}"
 
     # Check condition number is very high
-    cond = dfGB['diag_cond_xtx'].values[0]
+    cond = dfGB['diag_cond_xtx_test'].values[0]
     assert cond > 1e10, f"Expected high condition number, got {cond}"
 
     # Check function didn't crash (values are finite or NaN)
@@ -2008,14 +2008,14 @@ def test_v4_inf_nan_filtering():
     )
 
     # Check diagnostics exist
-    assert 'diag_n_total' in dfGB.columns
-    assert 'diag_n_filtered' in dfGB.columns
+    assert 'diag_n_total_test' in dfGB.columns
+    assert 'diag_n_filtered_test' in dfGB.columns
 
     # Each group should have filtered 1 row
-    assert all(dfGB['diag_n_filtered'] == 1)
+    assert all(dfGB['diag_n_filtered_test'] == 1)
 
     # All should have status OK (enough data remains)
-    assert all(dfGB['diag_status'] == 'OK')
+    assert all(dfGB['diag_status_test'] == 'OK')
 
     print("✅ test_v4_inf_nan_filtering passed")
 
@@ -2232,12 +2232,12 @@ def test_v4_singular_matrix_handling():
     )
 
     # Check status indicates problem
-    status = dfGB['diag_status'].values[0]
+    status = dfGB['diag_status_test'].values[0]
     assert status in ['ILL_CONDITIONED_RIDGED', 'SINGULAR_MATRIX'], \
         f"V4: Expected ill-conditioned status, got {status}"
 
     # Check condition number is very high
-    cond = dfGB['diag_cond_xtx'].values[0]
+    cond = dfGB['diag_cond_xtx_test'].values[0]
     assert cond > 1e10, f"V4: Expected high condition number, got {cond}"
 
     # Check function didn't crash (values are finite or NaN)