---
lang: ja-jp
breaks: true
---
# Python Pandas DataFrame `GROUP BY` `HAVING` と同じ処理を行う 2021-07-26

## BSSID毎の出現回数を算出する
appearances = 10
countBSSID = df[["BSSID", "SSID"]].groupby(["BSSID"]).filter(lambda x: x["SSID"].count() > appearances)
countBSSID = countBSSID.groupby(["BSSID"]).count().to_dict()["SSID"]
countBSSID = pd.DataFrame(list(countBSSID.items())).rename({0: "BSSID", 1: "COUNT"})
display(countBSSID)

	BSSID	COUNT
0	00:05:B5:4D:04:A8	13
1	00:05:B5:96:AB:D4	18
2	00:05:B5:E6:11:F9	39
3	00:1A:EB:C9:BF:00	11461
4	00:1A:EB:C9:BF:01	21935
...	...	...
217	FA:8F:CA:95:0E:DF	11
218	FA:B7:97:00:90:7B	25
219	FC:62:B9:75:3E:68	53
220	FC:87:43:01:55:31	78
221	FE:46:D7:20:10:9B	13
222 rows × 2 columns

###### tags: `Pandas` `Python` `DataFrame` `GROUP BY` `HAVING`