aki_prj23_transparenzregister/Jupyter/connection-counter.ipynb

In [54]:

from typing import Final

import numpy as np
import pandas as pd

In [1]:

from typing import Final

import numpy as np
import pandas as pd

number_of_entries = 100
number_of_contacts = 10
ids: Final = [_ for _ in range(number_of_entries)]
companies = pd.DataFrame(columns=[], index=pd.Index(ids, name="company_id"))
companies


id1 = (
    pd.Series(ids * number_of_contacts, name="Company 1")
    .sample(frac=0.7, random_state=42)
    .reset_index(drop=True)
)
id2 = (
    pd.Series(ids * number_of_contacts, name="Company 2")
    .sample(frac=0.7, random_state=43)
    .reset_index(drop=True)
)
connections = (
    pd.DataFrame(
        [
            id1,
            pd.Series(
                np.random.randint(0, 100, size=(max(len(id1), len(id2)))),
                name="Connection Weight",
            ),
            id2,
        ]
    )
    .T.dropna()
    .astype(int)
)
connections = connections.loc[(connections["Company 1"] != connections["Company 2"])]
connections

Out[1]:

<style scoped=""> .dataframe tbody tr th:only-of-type { vertical-align: middle; } .dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </style>

	Company 1	Connection Weight	Company 2
0	21	83	58
1	37	88	86
2	40	6	83
3	60	35	2
4	11	22	10
...	...	...	...
695	62	37	11
696	10	24	27
697	97	40	55
698	14	87	66
699	50	55	82

693 rows × 3 columns

In [69]:

id1 = (
    pd.Series(ids * number_of_contacts, name="Company 1")
    .sample(frac=0.7, random_state=42)
    .reset_index(drop=True)
)
id2 = (
    pd.Series(ids * number_of_contacts, name="Company 2")
    .sample(frac=0.7, random_state=43)
    .reset_index(drop=True)
)
connections = (
    pd.DataFrame(
        [
            id1,
            pd.Series(
                np.random.randint(0, 100, size=(max(len(id1), len(id2)))),
                name="Connection Weight",
            ),
            id2,
        ]
    )
    .T.dropna()
    .astype(int)
)
connections = connections.loc[(connections["Company 1"] != connections["Company 2"])]
connections

Out[69]:

<style scoped=""> .dataframe tbody tr th:only-of-type { vertical-align: middle; } .dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </style>

	Company 1	Connection Weight	Company 2
0	21	36	58
1	37	59	86
2	40	26	83
3	60	21	2
4	11	2	10
...	...	...	...
695	62	45	11
696	10	64	27
697	97	24	55
698	14	51	66
699	50	93	82

693 rows × 3 columns

In [73]:

connections[["Company 1", "Company 2"]].groupby("Company 1").count()

Out[73]:

<style scoped=""> .dataframe tbody tr th:only-of-type { vertical-align: middle; } .dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </style>

	Company 2
Company 1
0	6
1	6
2	5
3	9
4	7
...	...
95	7
96	8
97	7
98	6
99	8

100 rows × 1 columns

In [72]:

companies["Analysis-d0"] = 1
companies["Analysis-d1"] = connections[["Company 1", "Company 2"]].groupby("Company 1").count()
connection_sum = connections.join(connections.set_index("Company 2"), on=)
companies["Analysis-d1"] = connections[["Company 1", "Company 2"]].groupby("Company 1").count()
# for tiers in range(5):
companies

Out[72]:

<style scoped=""> .dataframe tbody tr th:only-of-type { vertical-align: middle; } .dataframe tbody tr th { vertical-align: top; } .dataframe thead th { text-align: right; } </style>

	Analysis-d0	Analysis-d1
company_id
0	1	6
1	1	6
2	1	5
3	1	9
4	1	7
...	...	...
95	1	7
96	1	8
97	1	7
98	1	6
99	1	8

100 rows × 2 columns

In [ ]:

companies

In [ ]: