Adding sklearn min. dependencies for all versions (#1022)

* Squashing commits

* All flow dependencies for sklearn>0.24 will change now

* Dep. string change only for OpenML>v0.11

Author: Neeratyoy

openml/extensions/sklearn/extension.py

@@ -211,6 +211,61 @@ def remove_all_in_parentheses(string: str) -> str:
 
         return short_name.format(pipeline)
 
+    @classmethod
+    def _min_dependency_str(cls, sklearn_version: str) -> str:
+        """ Returns a string containing the minimum dependencies for the sklearn version passed.
+
+        Parameters
+        ----------
+        sklearn_version : str
+            A version string of the xx.xx.xx
+
+        Returns
+        -------
+        str
+        """
+        openml_major_version = int(LooseVersion(openml.__version__).version[1])
+        # This explicit check is necessary to support existing entities on the OpenML servers
+        # that used the fixed dependency string (in the else block)
+        if openml_major_version > 11:
+            # OpenML v0.11 onwards supports sklearn>=0.24
+            # assumption: 0.24 onwards sklearn should contain a _min_dependencies.py file with
+            # variables declared for extracting minimum dependency for that version
+            if LooseVersion(sklearn_version) >= "0.24":
+                from sklearn import _min_dependencies as _mindep
+
+                dependency_list = {
+                    "numpy": "{}".format(_mindep.NUMPY_MIN_VERSION),
+                    "scipy": "{}".format(_mindep.SCIPY_MIN_VERSION),
+                    "joblib": "{}".format(_mindep.JOBLIB_MIN_VERSION),
+                    "threadpoolctl": "{}".format(_mindep.THREADPOOLCTL_MIN_VERSION),
+                }
+            elif LooseVersion(sklearn_version) >= "0.23":
+                dependency_list = {
+                    "numpy": "1.13.3",
+                    "scipy": "0.19.1",
+                    "joblib": "0.11",
+                    "threadpoolctl": "2.0.0",
+                }
+                if LooseVersion(sklearn_version).version[2] == 0:
+                    dependency_list.pop("threadpoolctl")
+            elif LooseVersion(sklearn_version) >= "0.21":
+                dependency_list = {"numpy": "1.11.0", "scipy": "0.17.0", "joblib": "0.11"}
+            elif LooseVersion(sklearn_version) >= "0.19":
+                dependency_list = {"numpy": "1.8.2", "scipy": "0.13.3"}
+            else:
+                dependency_list = {"numpy": "1.6.1", "scipy": "0.9"}
+        else:
+            # this is INCORRECT for sklearn versions >= 0.19 and < 0.24
+            # given that OpenML has existing flows uploaded with such dependency information,
+            # we change no behaviour for older sklearn version, however from 0.24 onwards
+            # the dependency list will be accurately updated for any flow uploaded to OpenML
+            dependency_list = {"numpy": "1.6.1", "scipy": "0.9"}
+
+        sklearn_dep = "sklearn=={}".format(sklearn_version)
+        dep_str = "\n".join(["{}>={}".format(k, v) for k, v in dependency_list.items()])
+        return "\n".join([sklearn_dep, dep_str])
+
     ################################################################################################
     # Methods for flow serialization and de-serialization
 
@@ -769,20 +824,13 @@ def _serialize_model(self, model: Any) -> OpenMLFlow:
             tags=tags,
             extension=self,
             language="English",
-            # TODO fill in dependencies!
             dependencies=dependencies,
         )
 
         return flow
 
     def _get_dependencies(self) -> str:
-        dependencies = "\n".join(
-            [
-                self._format_external_version("sklearn", sklearn.__version__,),
-                "numpy>=1.6.1",
-                "scipy>=0.9",
-            ]
-        )
+        dependencies = self._min_dependency_str(sklearn.__version__)
         return dependencies
 
     def _get_tags(self) -> List[str]:

tests/test_extensions/test_sklearn_extension/test_sklearn_extension.py

@@ -146,7 +146,7 @@ def test_serialize_model(self):
         fixture_short_name = "sklearn.DecisionTreeClassifier"
         # str obtained from self.extension._get_sklearn_description(model)
         fixture_description = "A decision tree classifier."
-        version_fixture = "sklearn==%s\nnumpy>=1.6.1\nscipy>=0.9" % sklearn.__version__
+        version_fixture = self.extension._min_dependency_str(sklearn.__version__)
 
         presort_val = "false" if LooseVersion(sklearn.__version__) < "0.22" else '"deprecated"'
         # min_impurity_decrease has been introduced in 0.20
@@ -227,7 +227,7 @@ def test_serialize_model_clustering(self):
         fixture_description = "K-Means clustering{}".format(
             "" if LooseVersion(sklearn.__version__) < "0.22" else "."
         )
-        version_fixture = "sklearn==%s\nnumpy>=1.6.1\nscipy>=0.9" % sklearn.__version__
+        version_fixture = self.extension._min_dependency_str(sklearn.__version__)
 
         n_jobs_val = "null" if LooseVersion(sklearn.__version__) < "0.23" else '"deprecated"'
         precomp_val = '"auto"' if LooseVersion(sklearn.__version__) < "0.23" else '"deprecated"'

tests/test_flows/test_flow_functions.py

@@ -343,7 +343,8 @@ def test_get_flow_reinstantiate_model_wrong_version(self):
             flow = openml.flows.get_flow(flow_id=flow, reinstantiate=True, strict_version=False)
             # ensure that a new flow was created
             assert flow.flow_id is None
-            assert "0.19.1" not in flow.dependencies
+            assert "sklearn==0.19.1" not in flow.dependencies
+            assert "sklearn>=0.19.1" not in flow.dependencies
 
     def test_get_flow_id(self):
         if self.long_version: