pheval 0.6.4__tar.gz → 0.6.6__tar.gz

{pheval-0.6.4 → pheval-0.6.6}/PKG-INFO

@@ -1,7 +1,8 @@
-Metadata-Version: 2.3
+Metadata-Version: 2.4
 Name: pheval
-Version: 0.6.4
+Version: 0.6.6
 Summary: 
+License-File: LICENSE
 Author: Yasemin Bridges
 Author-email: y.bridges@qmul.ac.uk
 Requires-Python: >=3.10,<4.0.0
@@ -10,6 +11,7 @@ Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
 Requires-Dist: class-resolver (>=0.4.2)
 Requires-Dist: click (>=8.1.3)
 Requires-Dist: deprecation (>=2.1.0)

{pheval-0.6.4 → pheval-0.6.6}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "pheval"
-version = "0.6.4"
+version = "0.6.6"
 description = ""
 authors = ["Yasemin Bridges <y.bridges@qmul.ac.uk>",
   "Julius Jacobsen <j.jacobsen@qmul.ac.uk>",

{pheval-0.6.4 → pheval-0.6.6}/src/pheval/analyse/benchmark.py

@@ -6,7 +6,7 @@ import duckdb
 import polars as pl
 
 from pheval.analyse.benchmark_db_manager import write_table
-from pheval.analyse.benchmark_output_type import BenchmarkOutputType, BenchmarkOutputTypeEnum
+from pheval.analyse.benchmark_output_type import BenchmarkOutputType, BenchmarkOutputTypeEnum, _get_unique_subset
 from pheval.analyse.binary_classification_curves import compute_curves
 from pheval.analyse.binary_classification_stats import compute_confusion_matrix
 from pheval.analyse.generate_plots import generate_plots
@@ -48,7 +48,7 @@ def scan_directory(run: RunConfig, benchmark_type: BenchmarkOutputType) -> pl.La
             else True
         )
         .sort("rank")
-        .unique(subset=["file_path", *benchmark_type.columns], keep="first")
+        .unique(subset=_get_unique_subset(benchmark_type), keep="first")
     )
 
 

{pheval-0.6.4 → pheval-0.6.6}/src/pheval/analyse/benchmark_output_type.py

@@ -42,3 +42,12 @@ class BenchmarkOutputTypeEnum(Enum):
         ["disease_identifier", "mondo_identifier"],
         "pheval_disease_results",
     )
+
+
+def _get_unique_subset(benchmark_type: BenchmarkOutputType) -> list[str]:
+    """
+    Determine which columns to use for uniqueness in benchmarking results.
+    """
+    if benchmark_type.prioritisation_type_string == "disease":
+        return ["file_path", "mondo_identifier"]
+    return ["file_path", *benchmark_type.columns]

{pheval-0.6.4 → pheval-0.6.6}/src/pheval/analyse/generate_rank_comparisons.py

@@ -1,7 +1,7 @@
 from itertools import combinations
 
 import polars as pl
-from duckdb.duckdb import DuckDBPyConnection
+from duckdb import DuckDBPyConnection
 
 from pheval.analyse.benchmark_db_manager import write_table
 from pheval.analyse.benchmark_output_type import BenchmarkOutputType