feat: split inventory

hakeem/core/utils/inventory.py

@@ -9,66 +9,11 @@ class _DEFAULT_RELIABILITY_BOUNDS:
 
 DEFAULT_RELIABILITY_BOUNDS = _DEFAULT_RELIABILITY_BOUNDS()
 
-"""
-The rest of the file contains utilities specific for the paper results reproduction.
-"""
-
-
-@dataclass(frozen=True)
-class Dataset:
-	path: str
-	alternatives: list
-	nbr_questions: int
-	nbr_voters: int
-
 
 @dataclass(frozen=True)
 class _COLUMNS:
-	interface: str = "Interface"
-	mechanism: str = "Mechanism"
 	question: str = "Question"
-	true_answer: str = "TrueAnswer"
-	answer: str = "Answer"
-	comments: str = "Comments"
 	voter: str = "Voter"
-	weight: str = "Weight"
 
 
 COLUMNS = _COLUMNS()
-
-
-ANIMALS_DATASET = Dataset(
-	"data/animals/raw.csv",
-	["Leopard", "Tiger", "Puma", "Jaguar", "Lion(ess)", "Cheetah"],
-	16,
-	110,
-)
-
-TEXTURES_DATASET = Dataset(
-	"data/textures/raw.csv",
-	["Gravel", "Grass", "Brick", "Wood", "Sand", "Cloth"],
-	16,
-	96,
-)
-
-LANGUAGE_DATASET = Dataset(
-	"data/languages/raw.csv",
-	[
-		"Hebrew",
-		"Russian",
-		"Japanese",
-		"Thai",
-		"Chinese",
-		"Tamil",
-		"Latin",
-		"Hindi",
-	],
-	25,
-	109,
-)
-
-DATASETS = {
-	"animals": ANIMALS_DATASET,
-	"textures": TEXTURES_DATASET,
-	"languages": LANGUAGE_DATASET,
-}

hakeem/paper_results/evaluation/accuracy.py

@@ -18,8 +18,8 @@
 	StandardApprovalAggregator,
 )
 from hakeem.core.aggregation.base import Aggregator
-from hakeem.core.utils.inventory import COLUMNS
 from hakeem.core.utils.utils import get_mean_confidence_interval
+from hakeem.paper_results.inventory import COLUMNS
 
 logging.basicConfig(
 	level=logging.INFO, format="'%(asctime)s - %(levelname)s - %(message)s'"

hakeem/paper_results/inventory.py

@@ -0,0 +1,61 @@
+from dataclasses import dataclass
+
+
+@dataclass(frozen=True)
+class Dataset:
+	path: str
+	alternatives: list
+	nbr_questions: int
+	nbr_voters: int
+
+
+@dataclass(frozen=True)
+class _COLUMNS:
+	interface: str = "Interface"
+	mechanism: str = "Mechanism"
+	question: str = "Question"
+	true_answer: str = "TrueAnswer"
+	answer: str = "Answer"
+	comments: str = "Comments"
+	voter: str = "Voter"
+	weight: str = "Weight"
+
+
+COLUMNS = _COLUMNS()
+
+
+ANIMALS_DATASET = Dataset(
+	"data/animals/raw.csv",
+	["Leopard", "Tiger", "Puma", "Jaguar", "Lion(ess)", "Cheetah"],
+	16,
+	110,
+)
+
+TEXTURES_DATASET = Dataset(
+	"data/textures/raw.csv",
+	["Gravel", "Grass", "Brick", "Wood", "Sand", "Cloth"],
+	16,
+	96,
+)
+
+LANGUAGE_DATASET = Dataset(
+	"data/languages/raw.csv",
+	[
+		"Hebrew",
+		"Russian",
+		"Japanese",
+		"Thai",
+		"Chinese",
+		"Tamil",
+		"Latin",
+		"Hindi",
+	],
+	25,
+	109,
+)
+
+DATASETS = {
+	"animals": ANIMALS_DATASET,
+	"textures": TEXTURES_DATASET,
+	"languages": LANGUAGE_DATASET,
+}

hakeem/paper_results/main.py

@@ -2,11 +2,11 @@
 
 import pandas as pd
 
-from hakeem.core.utils.inventory import COLUMNS, DATASETS
 from hakeem.paper_results.evaluation.accuracy import (
 	compare_methods,
 	plot_accuracies,
 )
+from hakeem.paper_results.inventory import COLUMNS, DATASETS
 
 
 def _process_dataset() -> None:

hakeem/paper_results/parsing/data_preparation.py

@@ -1,7 +1,7 @@
 import numpy as np
 import pandas as pd
 
-from hakeem.core.utils.inventory import COLUMNS, Dataset
+from hakeem.paper_results.inventory import COLUMNS, Dataset
 
 
 def _get_column_names(name: str, nbr_questions: int) -> list[str]:

tests/paper_results/evaluation/test_accuracy.py

@@ -10,8 +10,8 @@ def test_compare_methods() -> None:
 	# Given
 	import random
 
-	from hakeem.core.utils.inventory import COLUMNS, DATASETS
 	from hakeem.paper_results.evaluation.accuracy import compare_methods
+	from hakeem.paper_results.inventory import COLUMNS, DATASETS
 
 	random.seed(42)
 	dataset = DATASETS["animals"]

tests/paper_results/parsing/test_data_preparation.py

@@ -3,14 +3,14 @@
 import pandas as pd
 import pytest
 
-from hakeem.core.utils.inventory import COLUMNS
+from hakeem.paper_results.inventory import COLUMNS
 
 
 @pytest.mark.e2e
 @pytest.mark.parametrize("dataset", ["animals", "languages", "textures"])
 def test_prepare_data(dataset) -> None:
 	# With
-	from hakeem.core.utils.inventory import DATASETS
+	from hakeem.paper_results.inventory import DATASETS
 	from hakeem.paper_results.parsing.data_preparation import prepare_data
 
 	dataset = DATASETS[dataset]