maxframe 0.1.0b5__cp310-cp310-macosx_11_0_arm64.whl

maxframe/dataframe/reduction/prod.py

@@ -0,0 +1,76 @@
+# Copyright 1999-2024 Alibaba Group Holding Ltd.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import numpy as np
+
+from ... import opcodes
+from ...core import OutputType
+from .aggregation import where_function
+from .core import DataFrameReductionMixin, DataFrameReductionOperator
+
+
+class DataFrameProd(DataFrameReductionOperator, DataFrameReductionMixin):
+    _op_type_ = opcodes.PROD
+    _func_name = "prod"
+
+    @property
+    def is_atomic(self):
+        return self.min_count == 0
+
+    @classmethod
+    def get_reduction_callable(cls, op):
+        skipna, min_count = op.skipna, op.min_count
+
+        def prod(value):
+            if min_count == 0:
+                return value.prod(skipna=skipna)
+            else:
+                return where_function(
+                    value.count() >= min_count, value.prod(skipna=skipna), np.nan
+                )
+
+        return prod
+
+
+def prod_series(df, axis=None, skipna=True, level=None, min_count=0, method=None):
+    op = DataFrameProd(
+        axis=axis,
+        skipna=skipna,
+        level=level,
+        min_count=min_count,
+        output_types=[OutputType.scalar],
+        method=method,
+    )
+    return op(df)
+
+
+def prod_dataframe(
+    df,
+    axis=None,
+    skipna=True,
+    level=None,
+    min_count=0,
+    numeric_only=None,
+    method=None,
+):
+    op = DataFrameProd(
+        axis=axis,
+        skipna=skipna,
+        level=level,
+        min_count=min_count,
+        numeric_only=numeric_only,
+        output_types=[OutputType.series],
+        method=method,
+    )
+    return op(df)

maxframe/dataframe/reduction/reduction_size.py

@@ -0,0 +1,36 @@
+# Copyright 1999-2024 Alibaba Group Holding Ltd.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from ... import opcodes
+from ...core import OutputType
+from .core import DataFrameReductionMixin, DataFrameReductionOperator
+
+
+class DataFrameSize(DataFrameReductionOperator, DataFrameReductionMixin):
+    _op_type_ = opcodes.REDUCTION_SIZE
+    _func_name = "size"
+
+    @property
+    def is_atomic(self):
+        return True
+
+
+def size_series(df):
+    op = DataFrameSize(output_types=[OutputType.scalar])
+    return op(df)
+
+
+def size_dataframe(df):
+    op = DataFrameSize(output_types=[OutputType.series])
+    return op(df)

maxframe/dataframe/reduction/sem.py

@@ -0,0 +1,69 @@
+# Copyright 1999-2024 Alibaba Group Holding Ltd.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from ... import opcodes
+from ...core import OutputType
+from ...serialization.serializables import Int32Field
+from .core import DataFrameReductionMixin, DataFrameReductionOperator
+
+
+class DataFrameSem(DataFrameReductionOperator, DataFrameReductionMixin):
+    _op_type_ = opcodes.SEM
+    _func_name = "sem"
+
+    ddof = Int32Field("ddof", default=None)
+
+    @classmethod
+    def get_reduction_callable(cls, op: "DataFrameSem"):
+        skipna, ddof = op.skipna, op.ddof
+
+        def sem(x):
+            var = x.var(skipna=skipna, ddof=ddof)
+            cnt = x.count()
+            return (var / cnt) ** 0.5
+
+        return sem
+
+
+def sem_series(series, axis=None, skipna=True, level=None, ddof=1, method=None):
+    op = DataFrameSem(
+        axis=axis,
+        skipna=skipna,
+        level=level,
+        ddof=ddof,
+        output_types=[OutputType.scalar],
+        method=method,
+    )
+    return op(series)
+
+
+def sem_dataframe(
+    df,
+    axis=None,
+    skipna=True,
+    level=None,
+    ddof=1,
+    numeric_only=None,
+    method=None,
+):
+    op = DataFrameSem(
+        axis=axis,
+        skipna=skipna,
+        level=level,
+        ddof=ddof,
+        numeric_only=numeric_only,
+        output_types=[OutputType.series],
+        method=method,
+    )
+    return op(df)

maxframe/dataframe/reduction/skew.py

@@ -0,0 +1,89 @@
+# Copyright 1999-2024 Alibaba Group Holding Ltd.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import numpy as np
+
+from ... import opcodes
+from ...core import ENTITY_TYPE, OutputType
+from ...serialization.serializables import BoolField
+from .core import DataFrameReductionMixin, DataFrameReductionOperator
+
+
+class DataFrameSkew(DataFrameReductionOperator, DataFrameReductionMixin):
+    _op_type_ = opcodes.SKEW
+    _func_name = "skew"
+
+    bias = BoolField("bias", default=None)
+
+    @classmethod
+    def get_reduction_callable(cls, op: "DataFrameSkew"):
+        from .aggregation import where_function
+
+        skipna, bias = op.skipna, op.bias
+
+        def skew(x):
+            cnt = x.count()
+            mean = x.mean(skipna=skipna)
+            divided = (
+                (x**3).mean(skipna=skipna)
+                - 3 * (x**2).mean(skipna=skipna) * mean
+                + 2 * mean**3
+            )
+            var = x.var(skipna=skipna, ddof=0)
+            if isinstance(var, ENTITY_TYPE) or var > 0:
+                val = where_function(var > 0, divided / var**1.5, np.nan)
+            else:
+                val = np.nan
+            if not bias:
+                val = where_function(
+                    (var > 0) & (cnt > 2),
+                    val * ((cnt * (cnt - 1)) ** 0.5 / (cnt - 2)),
+                    np.nan,
+                )
+            return val
+
+        return skew
+
+
+def skew_series(df, axis=None, skipna=True, level=None, bias=False, method=None):
+    op = DataFrameSkew(
+        axis=axis,
+        skipna=skipna,
+        level=level,
+        bias=bias,
+        output_types=[OutputType.scalar],
+        method=method,
+    )
+    return op(df)
+
+
+def skew_dataframe(
+    df,
+    axis=None,
+    skipna=True,
+    level=None,
+    numeric_only=None,
+    bias=False,
+    method=None,
+):
+    op = DataFrameSkew(
+        axis=axis,
+        skipna=skipna,
+        level=level,
+        numeric_only=numeric_only,
+        bias=bias,
+        output_types=[OutputType.series],
+        method=method,
+    )
+    return op(df)

maxframe/dataframe/reduction/std.py

@@ -0,0 +1,53 @@
+# Copyright 1999-2024 Alibaba Group Holding Ltd.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from ...tensor.arithmetic import sqrt
+from .var import var_dataframe, var_series
+
+
+def std_dataframe(
+    df,
+    axis=None,
+    skipna=True,
+    level=None,
+    ddof=1,
+    numeric_only=None,
+    method=None,
+):
+    ret = sqrt(
+        var_dataframe(
+            df,
+            axis=axis,
+            skipna=skipna,
+            level=level,
+            ddof=ddof,
+            numeric_only=numeric_only,
+            method=method,
+        )
+    )
+    return ret
+
+
+def std_series(series, axis=None, skipna=True, level=None, ddof=1, method=None):
+    ret = sqrt(
+        var_series(
+            series,
+            axis=axis,
+            skipna=skipna,
+            level=level,
+            ddof=ddof,
+            method=method,
+        )
+    )
+    return ret

maxframe/dataframe/reduction/str_concat.py

@@ -0,0 +1,48 @@
+# Copyright 1999-2024 Alibaba Group Holding Ltd.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from ... import opcodes
+from ...core import OutputType
+from ...serialization.serializables import StringField
+from .core import DataFrameReductionMixin, DataFrameReductionOperator
+
+
+class DataFrameStrConcat(DataFrameReductionOperator, DataFrameReductionMixin):
+    _op_type_ = opcodes.STR_CONCAT
+    _func_name = "str_concat"
+
+    sep = StringField("sep", default=None)
+    na_rep = StringField("na_rep", default=None)
+
+    def get_reduction_args(self, axis=None):
+        return dict(sep=self.sep, na_rep=self.na_rep)
+
+    @property
+    def is_atomic(self):
+        return True
+
+    @classmethod
+    def get_reduction_callable(cls, op: "DataFrameStrConcat"):
+        sep, na_rep = op.sep, op.na_rep
+
+        def str_concat(obj):
+            return build_str_concat_object(obj, sep=sep, na_rep=na_rep)
+
+        return str_concat
+
+
+def build_str_concat_object(df, sep=None, na_rep=None):
+    output_type = OutputType.series if df.ndim == 2 else OutputType.scalar
+    op = DataFrameStrConcat(sep=sep, na_rep=na_rep, output_types=[output_type])
+    return op(df)

maxframe/dataframe/reduction/sum.py

@@ -0,0 +1,77 @@
+# Copyright 1999-2024 Alibaba Group Holding Ltd.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import numpy as np
+
+from ... import opcodes
+from ...core import OutputType
+from .core import DataFrameReductionMixin, DataFrameReductionOperator
+
+
+class DataFrameSum(DataFrameReductionOperator, DataFrameReductionMixin):
+    _op_type_ = opcodes.SUM
+    _func_name = "sum"
+
+    @property
+    def is_atomic(self):
+        return self.min_count == 0
+
+    @classmethod
+    def get_reduction_callable(cls, op):
+        from .aggregation import where_function
+
+        skipna, min_count = op.skipna, op.min_count
+
+        def sum_(value):
+            if min_count == 0:
+                return value.sum(skipna=skipna)
+            else:
+                return where_function(
+                    value.count() >= min_count, value.sum(skipna=skipna), np.nan
+                )
+
+        return sum_
+
+
+def sum_series(df, axis=None, skipna=True, level=None, min_count=0, method=None):
+    op = DataFrameSum(
+        axis=axis,
+        skipna=skipna,
+        level=level,
+        min_count=min_count,
+        output_types=[OutputType.scalar],
+        method=method,
+    )
+    return op(df)
+
+
+def sum_dataframe(
+    df,
+    axis=None,
+    skipna=True,
+    level=None,
+    min_count=0,
+    numeric_only=None,
+    method=None,
+):
+    op = DataFrameSum(
+        axis=axis,
+        skipna=skipna,
+        level=level,
+        min_count=min_count,
+        numeric_only=numeric_only,
+        output_types=[OutputType.series],
+        method=method,
+    )
+    return op(df)

maxframe/dataframe/reduction/tests/__init__.py

@@ -0,0 +1,13 @@
+# Copyright 1999-2024 Alibaba Group Holding Ltd.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.