to preserve our scalar values 141 """ --> 142 return self.to_numpy(dtype=dtype) 143 144 def __arrow_array__(self, type=None): /pandas/pandas/core/arrays/masked.py in to_numpy(self, dtype, copy, na_value) datetime64[ns, US/Eastern] i datetime64[ns] dtype: object 3.1.12 Parquet New in version 0.21.0. Apache Parquet provides a partitioned binary columnar serialization for data frames. It is designed to make _metadata, __hash__, and changed the default definition of __eq__. For interaction with Apache Arrow (pyarrow), a __from_arrow__ method can be implemented: this method receives a pyarrow Array or ChunkedArray

. . . . . . . . . . . . . . . . . . . . . . . . . 2381 4.6.1 Storing pandas DataFrame objects in Apache Parquet format . . . . . . . . . . . . . . . . . 2381 4.7 Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2385 4.8.3 Apache Arrow interoperability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2386 4.8.4 datetime64[ns, US/Eastern] i datetime64[ns] dtype: object 2.1.12 Parquet New in version 0.21.0. Apache Parquet provides a partitioned binary columnar serialization for data frames. It is designed to make

. . . . . . . . . . . . . . . . . . . . . . . . . 2377 4.6.1 Storing pandas DataFrame objects in Apache Parquet format . . . . . . . . . . . . . . . . . 2377 4.7 Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2381 4.8.3 Apache Arrow interoperability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2382 4.8.4 datetime64[ns, US/Eastern] i datetime64[ns] dtype: object 2.1.12 Parquet New in version 0.21.0. Apache Parquet provides a partitioned binary columnar serialization for data frames. It is designed to make

. . . . . . . . . . . . . . . . . . . . . . . . . 2476 4.6.1 Storing pandas DataFrame objects in Apache Parquet format . . . . . . . . . . . . . . . . . 2476 4.7 Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2480 4.8.3 Apache Arrow interoperability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2480 4.8.4 g datetime64[ns] h datetime64[ns, US/Eastern] i datetime64[ns] dtype: object 2.4.12 Parquet Apache Parquet provides a partitioned binary columnar serialization for data frames. It is designed to make

. . . . . . . . . . . . . . . . . . . . . . . . . 2476 4.6.1 Storing pandas DataFrame objects in Apache Parquet format . . . . . . . . . . . . . . . . . 2476 4.7 Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2480 4.8.3 Apache Arrow interoperability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2480 4.8.4 g datetime64[ns] h datetime64[ns, US/Eastern] i datetime64[ns] dtype: object 2.4.12 Parquet Apache Parquet provides a partitioned binary columnar serialization for data frames. It is designed to make

. . . . . . . . . . . . . . . . . . . . . . . . . 2367 5.6.1 Storing pandas DataFrame objects in Apache Parquet format . . . . . . . . . . . . . . . . . 2367 5.7 Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2371 5.8.3 Apache Arrow interoperability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2372 5.8.4 datetime64[ns, US/Eastern] i datetime64[ns] dtype: object 3.1.12 Parquet New in version 0.21.0. Apache Parquet provides a partitioned binary columnar serialization for data frames. It is designed to make

. . . . . . . . . . . . . . . . . . . . . . . . 2674 4.11.1 Storing pandas DataFrame objects in Apache Parquet format . . . . . . . . . . . . . . . . . 2674 4.12 Policies . . . . . . . . . . . . . . missing value handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2678 4.13.4 Apache Arrow interoperability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2678 4.13 removed in version 1.0.0. It is recommended to use pickle instead. Alternatively, you can also the Arrow IPC serialization format for on-the-wire transmission of pandas objects. For documentation on pyarrow

. . . . . . . . . . . . . . . . . . . . . . . . 2753 4.11.1 Storing pandas DataFrame objects in Apache Parquet format . . . . . . . . . . . . . . . . . 2753 4.12 Policies . . . . . . . . . . . . . . missing value handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2757 4.13.4 Apache Arrow interoperability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2757 4.13 removed in version 1.0.0. It is recommended to use pickle instead. Alternatively, you can also the Arrow IPC serialization format for on-the-wire transmission of pandas objects. For documentation on pyarrow

. . . . . . . . . . . . . . . . . . . . . . . . 2753 4.11.1 Storing pandas DataFrame objects in Apache Parquet format . . . . . . . . . . . . . . . . . 2753 4.12 Policies . . . . . . . . . . . . . . missing value handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2757 4.13.4 Apache Arrow interoperability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2757 4.13 removed in version 1.0.0. It is recommended to use pickle instead. Alternatively, you can also the Arrow IPC serialization format for on-the-wire transmission of pandas objects. For documentation on pyarrow

. . . . . . . . . . . . . . . . . . . . . . . . . 2531 4.8.1 Storing pandas DataFrame objects in Apache Parquet format . . . . . . . . . . . . . . . . . 2531 4.9 Policies . . . . . . . . . . . . . . missing value handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2536 4.10.4 Apache Arrow interoperability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2536 4.10 g datetime64[ns] h datetime64[ns, US/Eastern] i datetime64[ns] dtype: object 2.4.12 Parquet Apache Parquet provides a partitioned binary columnar serialization for data frames. It is designed to make