FIX-#2369: Update pandas version to 1.1.4 (#2371)

Signed-off-by: Igoshev, Yaroslav <yaroslav.igoshev@intel.com>

environment.yml

@@ -2,7 +2,7 @@ name: modin
 channels:
   - conda-forge
 dependencies:
-  - pandas==1.1.3
+  - pandas==1.1.4
   - numpy
   - pyarrow==1.0
   - dask[complete]>=2.12.0,<=2.19.0

modin/engines/base/io/file_reader.py

@@ -137,10 +137,10 @@ def file_exists(cls, file_path):
         return os.path.exists(file_path)
 
     @classmethod
-    def deploy(cls, func, args, num_return_vals):
+    def deploy(cls, func, args, num_returns):
         raise NotImplementedError(NOT_IMPLEMENTED_MESSAGE)
 
-    def parse(self, func, args, num_return_vals):
+    def parse(self, func, args, num_returns):
         raise NotImplementedError(NOT_IMPLEMENTED_MESSAGE)
 
     @classmethod

modin/engines/dask/task_wrapper.py

@@ -16,12 +16,12 @@
 
 class DaskTask:
     @classmethod
-    def deploy(cls, func, num_return_vals, kwargs):
+    def deploy(cls, func, num_returns, kwargs):
         client = _get_global_client()
         remote_task_future = client.submit(func, **kwargs)
         return [
             client.submit(lambda l, i: l[i], remote_task_future, i)
-            for i in range(num_return_vals)
+            for i in range(num_returns)
         ]
 
     @classmethod

modin/experimental/engines/pandas_on_ray/io_exp.py

@@ -148,7 +148,7 @@ def read_sql(
                     columns,
                     chunksize,
                 ),
-                num_return_vals=num_splits + 1,
+                num_returns=num_splits + 1,
             )
             partition_ids.append(
                 [PandasOnRayFramePartition(obj) for obj in partition_id[:-1]]

modin/experimental/engines/pyarrow_on_ray/frame/axis_partition.py

@@ -46,15 +46,15 @@ def apply(self, func, num_splits=None, other_axis_partition=None, **kwargs):
                 for obj in deploy_ray_func_between_two_axis_partitions._remote(
                     args=(self.axis, func, num_splits, len(self.list_of_blocks), kwargs)
                     + tuple(self.list_of_blocks + other_axis_partition.list_of_blocks),
-                    num_return_vals=num_splits,
+                    num_returns=num_splits,
                 )
             ]
 
         args = [self.axis, func, num_splits, kwargs]
         args.extend(self.list_of_blocks)
         return [
             PyarrowOnRayFramePartition(obj)
-            for obj in deploy_ray_axis_func._remote(args, num_return_vals=num_splits)
+            for obj in deploy_ray_axis_func._remote(args, num_returns=num_splits)
         ]
 
     def shuffle(self, func, num_splits=None, **kwargs):
@@ -74,7 +74,7 @@ def shuffle(self, func, num_splits=None, **kwargs):
         args.extend(self.list_of_blocks)
         return [
             PyarrowOnRayFramePartition(obj)
-            for obj in deploy_ray_axis_func._remote(args, num_return_vals=num_splits)
+            for obj in deploy_ray_axis_func._remote(args, num_returns=num_splits)
         ]
 
 

modin/pandas/__init__.py

@@ -13,7 +13,7 @@
 
 import pandas
 
-__pandas_version__ = "1.1.3"
+__pandas_version__ = "1.1.4"
 
 if pandas.__version__ != __pandas_version__:
     import warnings

modin/pandas/base.py

@@ -488,9 +488,6 @@ def add(self, other, axis="columns", level=None, fill_value=None):
         )
 
     def aggregate(self, func=None, axis=0, *args, **kwargs):
-        warnings.warn(
-            "Modin index may not match pandas index due to pandas issue pandas-dev/pandas#36189."
-        )
         axis = self._get_axis_number(axis)
         result = None
 
@@ -686,9 +683,6 @@ def apply(
         args=(),
         **kwds,
     ):
-        warnings.warn(
-            "Modin index may not match pandas index due to pandas issue pandas-dev/pandas#36189."
-        )
         axis = self._get_axis_number(axis)
         ErrorMessage.non_verified_udf()
         if isinstance(func, str):

modin/pandas/test/dataframe/test_binary.py

@@ -135,7 +135,7 @@ def test_math_alias(math_op, alias):
     assert getattr(pd.DataFrame, math_op) == getattr(pd.DataFrame, alias)
 
 
-@pytest.mark.parametrize("other", ["as_left", 4, 4.0])
+@pytest.mark.parametrize("other", ["as_left", 4, 4.0, "a"])
 @pytest.mark.parametrize("op", ["eq", "ge", "gt", "le", "lt", "ne"])
 @pytest.mark.parametrize("data", test_data_values, ids=test_data_keys)
 def test_comparison(data, op, other):
@@ -145,20 +145,6 @@ def test_comparison(data, op, other):
     )
 
 
-@pytest.mark.xfail_backends(
-    ["BaseOnPython"],
-    reason="Test is failing because of mismathing of thrown exceptions. See pandas issue #36377",
-)
-@pytest.mark.parametrize("other", ["a"])
-@pytest.mark.parametrize("op", ["ge", "gt", "le", "lt", "eq", "ne"])
-@pytest.mark.parametrize("data", test_data_values, ids=test_data_keys)
-def test_comparison_except(data, op, other):
-    eval_general(
-        *create_test_dfs(data),
-        lambda df: getattr(df, op)(other),
-    )
-
-
 @pytest.mark.parametrize("op", ["eq", "ge", "gt", "le", "lt", "ne"])
 @pytest.mark.parametrize("data", test_data_values, ids=test_data_keys)
 def test_multi_level_comparison(data, op):

modin/pandas/test/dataframe/test_udf.py

@@ -49,16 +49,10 @@
 )
 @pytest.mark.parametrize("op", ["agg", "apply"])
 def test_agg_apply(axis, func, op):
-    # AssertionError may be arisen in case of
-    # mismathing of index/columns in Modin and pandas.
-    # See details in pandas issue 36189.
-    try:
-        eval_general(
-            *create_test_dfs(test_data["float_nan_data"]),
-            lambda df: getattr(df, op)(func, axis),
-        )
-    except AssertionError:
-        pass
+    eval_general(
+        *create_test_dfs(test_data["float_nan_data"]),
+        lambda df: getattr(df, op)(func, axis),
+    )
 
 
 @pytest.mark.parametrize("axis", ["rows", "columns"])
@@ -69,16 +63,10 @@ def test_agg_apply(axis, func, op):
 )
 @pytest.mark.parametrize("op", ["agg", "apply"])
 def test_agg_apply_axis_names(axis, func, op):
-    # AssertionError may be arisen in case of
-    # mismathing of index/columns in Modin and pandas.
-    # See details in pandas issue 36189.
-    try:
-        eval_general(
-            *create_test_dfs(test_data["int_data"]),
-            lambda df: getattr(df, op)(func, axis),
-        )
-    except AssertionError:
-        pass
+    eval_general(
+        *create_test_dfs(test_data["int_data"]),
+        lambda df: getattr(df, op)(func, axis),
+    )
 
 
 def test_aggregate_alias():

modin/pandas/test/test_series.py

@@ -593,16 +593,10 @@ def test_add_suffix(data):
 @pytest.mark.parametrize("data", test_data_values, ids=test_data_keys)
 @pytest.mark.parametrize("func", agg_func_values, ids=agg_func_keys)
 def test_agg(data, func):
-    # AssertionError may be arisen in case of
-    # mismathing of index/columns in Modin and pandas.
-    # See details in pandas issue 36189.
-    try:
-        eval_general(
-            *create_test_series(data),
-            lambda df: df.agg(func),
-        )
-    except AssertionError:
-        pass
+    eval_general(
+        *create_test_series(data),
+        lambda df: df.agg(func),
+    )
 
 
 @pytest.mark.parametrize("data", test_data_values, ids=test_data_keys)
@@ -624,16 +618,10 @@ def test_agg_numeric(request, data, func):
         request.node.name, numeric_dfs
     ):
         axis = 0
-        # AssertionError may be arisen in case of
-        # mismathing of index/columns in Modin and pandas.
-        # See details in pandas issue 36189.
-        try:
-            eval_general(
-                *create_test_series(data),
-                lambda df: df.agg(func, axis),
-            )
-        except AssertionError:
-            pass
+        eval_general(
+            *create_test_series(data),
+            lambda df: df.agg(func, axis),
+        )
 
 
 @pytest.mark.parametrize("data", test_data_values, ids=test_data_keys)
@@ -656,16 +644,10 @@ def test_agg_numeric_except(request, data, func):
 @pytest.mark.parametrize("func", agg_func_values, ids=agg_func_keys)
 def test_aggregate(data, func):
     axis = 0
-    # AssertionError may be arisen in case of
-    # mismathing of index/columns in Modin and pandas.
-    # See details in pandas issue 36189.
-    try:
-        eval_general(
-            *create_test_series(data),
-            lambda df: df.aggregate(func, axis),
-        )
-    except AssertionError:
-        pass
+    eval_general(
+        *create_test_series(data),
+        lambda df: df.aggregate(func, axis),
+    )
 
 
 @pytest.mark.parametrize("data", test_data_values, ids=test_data_keys)
@@ -688,16 +670,10 @@ def test_aggregate_numeric(request, data, func):
         request.node.name, numeric_dfs
     ):
         axis = 0
-        # AssertionError may be arisen in case of
-        # mismathing of index/columns in Modin and pandas.
-        # See details in pandas issue 36189.
-        try:
-            eval_general(
-                *create_test_series(data),
-                lambda df: df.agg(func, axis),
-            )
-        except AssertionError:
-            pass
+        eval_general(
+            *create_test_series(data),
+            lambda df: df.agg(func, axis),
+        )
 
 
 @pytest.mark.parametrize("data", test_data_values, ids=test_data_keys)
@@ -823,16 +799,10 @@ def test_append(data):
 @pytest.mark.parametrize("data", test_data_values, ids=test_data_keys)
 @pytest.mark.parametrize("func", agg_func_values, ids=agg_func_keys)
 def test_apply(data, func):
-    # AssertionError may be arisen in case of
-    # mismathing of index/columns in Modin and pandas.
-    # See details in pandas issue 36189.
-    try:
-        eval_general(
-            *create_test_series(data),
-            lambda df: df.apply(func),
-        )
-    except AssertionError:
-        pass
+    eval_general(
+        *create_test_series(data),
+        lambda df: df.apply(func),
+    )
 
 
 @pytest.mark.parametrize("data", test_data_values, ids=test_data_keys)
@@ -871,16 +841,10 @@ def test_apply_external_lib():
 @pytest.mark.parametrize("func", agg_func_values, ids=agg_func_keys)
 def test_apply_numeric(request, data, func):
     if name_contains(request.node.name, numeric_dfs):
-        # AssertionError may be arisen in case of
-        # mismathing of index/columns in Modin and pandas.
-        # See details in pandas issue 36189.
-        try:
-            eval_general(
-                *create_test_series(data),
-                lambda df: df.apply(func),
-            )
-        except AssertionError:
-            pass
+        eval_general(
+            *create_test_series(data),
+            lambda df: df.apply(func),
+        )
 
 
 @pytest.mark.parametrize("data", test_data_values, ids=test_data_keys)

requirements.txt

@@ -1,4 +1,4 @@
-pandas==1.1.3
+pandas==1.1.4
 numpy
 pyarrow==1.0
 dask[complete]>=2.12.0,<=2.19.0

requirements/env_omnisci.yml

@@ -3,7 +3,7 @@ channels:
   - intel/label/modin
   - conda-forge
 dependencies:
-  - pandas==1.1.3
+  - pandas==1.1.4
   - pyarrow==1.0
   - numpy
   - pip

setup.py

@@ -55,7 +55,7 @@ def is_pure(self):
     url="https://github.com/modin-project/modin",
     long_description=long_description,
     long_description_content_type="text/markdown",
-    install_requires=["pandas==1.1.3", "packaging"],
+    install_requires=["pandas==1.1.4", "packaging"],
     extras_require={
         # can be installed by pip install modin[dask]
         "dask": dask_deps,