From 9e3388bee2e61ab90f38377701a44a1080cc7a75 Mon Sep 17 00:00:00 2001
From: Shelly Chahar <shellychahar57@gmail.com>
Date: Tue, 6 Oct 2026 12:47:17 +0530
Subject: [PATCH] fix(tasks): infer task_type string from TaskType in
 create_task (#1766)

- Add type_str property to TaskType enum mapping members to OpenML task strings

- Resolve TODO in create_task() by passing task_type.type_str instead of hardcoded 'None'

- Default task_type to task_type_id.type_str in OpenMLTask.__init__ if not provided

- Add unit tests for task_type assignment and TaskType.type_str

Signed-off-by: Shelly Chahar <shellychahar57@gmail.com>
---
 openml/tasks/functions.py               |  2 +-
 openml/tasks/task.py                    | 18 ++++++++-
 tests/test_tasks/test_task_functions.py | 53 +++++++++++++++++++++++++
 3 files changed, 71 insertions(+), 2 deletions(-)

diff --git a/openml/tasks/functions.py b/openml/tasks/functions.py
index e123b8638..ead8a8a5c 100644
--- a/openml/tasks/functions.py
+++ b/openml/tasks/functions.py
@@ -572,7 +572,7 @@ def create_task(
     return task_cls(
         task_id=None,
         task_type_id=task_type,
-        task_type="None",  # TODO: refactor to get task type string from ID.
+        task_type=task_type.type_str,
         data_set_id=dataset_id,
         target_name=target_name,  # type: ignore
         estimation_procedure_id=estimation_procedure_id,
diff --git a/openml/tasks/task.py b/openml/tasks/task.py
index ab3cb3da4..398ac8235 100644
--- a/openml/tasks/task.py
+++ b/openml/tasks/task.py
@@ -41,6 +41,22 @@ class TaskType(Enum):
     SUBGROUP_DISCOVERY = 8
     MULTITASK_REGRESSION = 9
 
+    @property
+    def type_str(self) -> str:
+        """Return the OpenML string representation of the task type."""
+        mapping = {
+            TaskType.SUPERVISED_CLASSIFICATION: "Supervised Classification",
+            TaskType.SUPERVISED_REGRESSION: "Supervised Regression",
+            TaskType.LEARNING_CURVE: "Learning Curve",
+            TaskType.SUPERVISED_DATASTREAM_CLASSIFICATION: "Supervised Data Stream Classification",
+            TaskType.CLUSTERING: "Clustering",
+            TaskType.MACHINE_LEARNING_CHALLENGE: "Machine Learning Challenge",
+            TaskType.SURVIVAL_ANALYSIS: "Survival Analysis",
+            TaskType.SUBGROUP_DISCOVERY: "Subgroup Discovery",
+            TaskType.MULTITASK_REGRESSION: "Multi-task Regression",
+        }
+        return mapping[self]
+
 
 class _EstimationProcedure(TypedDict):
     type: str | None
@@ -90,7 +106,7 @@ def __init__(  # noqa: PLR0913
     ):
         self.task_id = int(task_id) if task_id is not None else None
         self.task_type_id = task_type_id
-        self.task_type = task_type
+        self.task_type = task_type if task_type is not None else task_type_id.type_str
         self.dataset_id = int(data_set_id)
         self.target_name = target_name
         resolved_estimation_procedure_id = self._resolve_estimation_procedure_id(
diff --git a/tests/test_tasks/test_task_functions.py b/tests/test_tasks/test_task_functions.py
index bf2fcfeae..dd73f7ce7 100644
--- a/tests/test_tasks/test_task_functions.py
+++ b/tests/test_tasks/test_task_functions.py
@@ -315,3 +315,56 @@ def test_delete_unknown_task(mock_delete, test_files_directory, test_server_v1,
     task_url = test_server_v1 + "task/9999999"
     assert task_url == mock_delete.call_args.args[0]
     assert test_apikey_v1 == mock_delete.call_args.kwargs.get("params", {}).get("api_key")
+
+
+def test_create_task_sets_task_type() -> None:
+    task = openml.tasks.create_task(
+        TaskType.SUPERVISED_CLASSIFICATION,
+        dataset_id=1,
+        target_name="class",
+        estimation_procedure_id=1,
+    )
+    assert task.task_type == "Supervised Classification"
+    assert task.task_type_id == TaskType.SUPERVISED_CLASSIFICATION
+
+    task_reg = openml.tasks.create_task(
+        TaskType.SUPERVISED_REGRESSION,
+        dataset_id=1,
+        target_name="target",
+        estimation_procedure_id=7,
+    )
+    assert task_reg.task_type == "Supervised Regression"
+    assert task_reg.task_type_id == TaskType.SUPERVISED_REGRESSION
+
+    task_cluster = openml.tasks.create_task(
+        TaskType.CLUSTERING,
+        dataset_id=1,
+        estimation_procedure_id=17,
+    )
+    assert task_cluster.task_type == "Clustering"
+    assert task_cluster.task_type_id == TaskType.CLUSTERING
+
+    task_lc = openml.tasks.create_task(
+        TaskType.LEARNING_CURVE,
+        dataset_id=1,
+        target_name="class",
+        estimation_procedure_id=13,
+    )
+    assert task_lc.task_type == "Learning Curve"
+    assert task_lc.task_type_id == TaskType.LEARNING_CURVE
+
+
+def test_task_type_str() -> None:
+    expected = {
+        TaskType.SUPERVISED_CLASSIFICATION: "Supervised Classification",
+        TaskType.SUPERVISED_REGRESSION: "Supervised Regression",
+        TaskType.LEARNING_CURVE: "Learning Curve",
+        TaskType.SUPERVISED_DATASTREAM_CLASSIFICATION: "Supervised Data Stream Classification",
+        TaskType.CLUSTERING: "Clustering",
+        TaskType.MACHINE_LEARNING_CHALLENGE: "Machine Learning Challenge",
+        TaskType.SURVIVAL_ANALYSIS: "Survival Analysis",
+        TaskType.SUBGROUP_DISCOVERY: "Subgroup Discovery",
+        TaskType.MULTITASK_REGRESSION: "Multi-task Regression",
+    }
+    for task_type, expected_str in expected.items():
+        assert task_type.type_str == expected_str