[docs] [ci] skip scikit-learn 1.7.1 in docs build, make tests compatible with...

[docs] [ci] skip scikit-learn 1.7.1 in docs build, make tests compatible with pandas 3.0, use new Azure DevOps pool (#6979)

.vsts-ci.yml

@@ -53,7 +53,7 @@ jobs:
   # Maintenance #
   ###############
   - job: Maintenance
-    pool: lgb_agent_pool_ado
+    pool: lightgbm_agent_pool_ado
     container: ubuntu-latest
     # routine maintenance (like periodically deleting old files),
     # to be run on 1 random CI runner in the self-hosted pool each runner
@@ -90,7 +90,7 @@ jobs:
       SETUP_CONDA: 'false'
       OS_NAME: 'linux'
       PRODUCES_ARTIFACTS: 'true'
-    pool: lgb_agent_pool_ado
+    pool: lightgbm_agent_pool_ado
     container: linux-artifact-builder
     strategy:
       matrix:
@@ -157,7 +157,7 @@ jobs:
       IN_UBUNTU_BASE_CONTAINER: 'true'
       OS_NAME: 'linux'
       SETUP_CONDA: 'true'
-    pool: lgb_agent_pool_ado
+    pool: lightgbm_agent_pool_ado
     container: ubuntu-latest
     strategy:
       matrix:

docs/env.yml

@@ -13,6 +13,7 @@ dependencies:
   - r-matrix=1.6_5
   - r-pkgdown=2.1.1
   - r-roxygen2=7.3.2
-  - scikit-learn>=1.6.1
+  # skipping scikit-learn 1.7.1 because of the problems described in https://github.com/microsoft/LightGBM/issues/6978
+  - scikit-learn>=1.6.1,!=1.7.1
   - sphinx>=8.1.3
   - sphinx_rtd_theme>=3.0.1

tests/python_package_test/test_basic.py

@@ -678,10 +678,22 @@ def test_list_to_1d_numpy(collection, dtype, rng):
     custom_name = "my_custom_variable"
 
     if collection.startswith("pd"):
-        if not PANDAS_INSTALLED:
-            pytest.skip("pandas is not installed")
-        else:
-            y = pd_Series(y)
+        pd = pytest.importorskip("pandas")
+        y = pd_Series(y)
+        if pd.api.types.is_object_dtype(y):
+            with pytest.raises(
+                ValueError,
+                match=r"pandas dtypes must be int, float or bool\.\nFields with bad pandas dtypes: 0: object",
+            ):
+                lgb.basic._list_to_1d_numpy(y, dtype=np.float32, name=custom_name)
+            return
+        elif pd.api.types.is_string_dtype(y):
+            with pytest.raises(
+                ValueError, match=r"pandas dtypes must be int, float or bool\.\nFields with bad pandas dtypes: 0: str"
+            ):
+                lgb.basic._list_to_1d_numpy(y, dtype=np.float32, name=custom_name)
+            return
+
     if isinstance(y, np.ndarray) and len(y.shape) == 2:
         with pytest.warns(UserWarning, match="column-vector"):
             lgb.basic._list_to_1d_numpy(y, dtype=np.float32, name=custom_name)
@@ -694,12 +706,7 @@ def test_list_to_1d_numpy(collection, dtype, rng):
         with pytest.raises(TypeError, match=err_msg):
             lgb.basic._list_to_1d_numpy(y, dtype=np.float32, name=custom_name)
         return
-    elif isinstance(y, pd_Series) and y.dtype == object:
-        with pytest.raises(
-            ValueError, match=r"pandas dtypes must be int, float or bool\.\nFields with bad pandas dtypes: 0: object"
-        ):
-            lgb.basic._list_to_1d_numpy(y, dtype=np.float32, name=custom_name)
-        return
+
     result = lgb.basic._list_to_1d_numpy(y, dtype=dtype, name=custom_name)
     assert result.size == 10
     assert result.dtype == dtype