hierarchy_attribute_evaluator
Evaluation utilities for the SNOMED CT attribute extraction pipeline.
Public API
build_prediction_rows — flatten pipeline output into row-wise predictions. evaluate_results — full outer join evaluation producing P/R/F1.
build_prediction_rows(results, hierarchy_settings)
Extract prediction rows from pipeline results for evaluation.
Handles regular attributes (single dict, list of dicts) and paired
interprets_interpretation structures via :func:split_interprets_pairs.
Each row carries relationship_groups: the SNOMED role group number(s) of the
attribute value, ;-separated (a value shared by several groups appears once).
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
results
|
list[dict]
|
Pipeline result dicts from |
required |
hierarchy_settings
|
HierarchySettings
|
Pipeline configuration. |
required |
Returns:
| Type | Description |
|---|---|
list[dict]
|
List of flat dicts ready for |
Source code in src/ariadne/evaluation/hierarchy_attribute_evaluator.py
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 | |
evaluate_results(results, gs_path, output_dir, hierarchy_settings)
Produce a combined evaluation table (full outer join of GS and predictions).
Columns
source_key, source_concept_id, source_concept_code, source_vocabulary_id,
source_term, attribute_category,
gs_target_concept_id, gs_target_concept_code, gs_target_concept_name,
predicted_target_concept_id, predicted_target_concept_code, predicted_target_concept_name,
matched, status (match / missed / extra).
Summary statistics are printed and the combined table is saved to CSV.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
results
|
list[dict]
|
List of pipeline result dicts from |
required |
gs_path
|
str
|
Path to the gold-standard CSV. |
required |
output_dir
|
Path
|
Path wherhe the output CSV should be saved. |
required |
hierarchy_settings
|
HierarchySettings
|
Pipeline configuration (see :func: |
required |
Returns:
| Type | Description |
|---|---|
DataFrame
|
Combined evaluation DataFrame. |
Source code in src/ariadne/evaluation/hierarchy_attribute_evaluator.py
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 | |