Pandas Integration

Graviti SDK provides converting methods to pandas DataFrame and Series.

Graviti to Pandas

Graviti SDK provides DataFrame.to_pandas() and Series.to_pandas() methods to convert graviti DataFrame and Series to pandas.

>>> from graviti import DataFrame

>>> df = DataFrame(
...     [
...         {
...             "A": i,
...             "B": f"data{i}",
...             "C": bool(i % 2),
...         }
...         for i in range(10)
...     ]
... )
>>> pandas_df = df.to_pandas()
>>> pandas_df
   A      B      C
0  0  data0  False
1  1  data1   True
2  2  data2  False
3  3  data3   True
4  4  data4  False
5  5  data5   True
6  6  data6  False
7  7  data7   True
8  8  data8  False
9  9  data9   True

>>> type(pandas_df)
pandas.core.frame.DataFrame

>>> from graviti import Series

>>> series = Series(range(10))
>>> pandas_series = series.to_pandas()
>>> pandas_series
0    0
1    1
2    2
3    3
4    4
5    5
6    6
7    7
8    8
9    9
dtype: int64
>>> type(pandas_series)
pandas.core.series.Series

Pandas to Graviti

Graviti SDK provides DataFrame.from_pandas() and Series.from_pandas() methods to convert pandas DataFrame and Series to graviti.

>>> import pandas as pd
>>> from graviti import DataFrame

>>> pandas_df = pd.DataFrame(
...     [
...         {
...             "A": i,
...             "B": f"data{i}",
...             "C": bool(i % 2),
...         }
...         for i in range(10)
...     ]
... )
>>> df = DataFrame.from_pandas(pandas_df)
>>> df
   A      B      C
0  0  data0  False
1  1  data1   True
2  2  data2  False
3  3  data3   True
4  4  data4  False
5  5  data5   True
6  6  data6  False
7  7  data7   True
8  8  data8  False
9  9  data9   True

>>> type(df)
graviti.dataframe.frame.DataFrame

>>> import pandas as pd
>>> from graviti import Series

>>> pandas_series = pd.Series(range(10))
>>> series = Series.from_pandas(pandas_series)
>>> series
0  0
1  1
2  2
3  3
4  4
5  5
6  6
7  7
8  8
9  9
>>> type(series)
graviti.dataframe.column.series.NumberSeries