cogs package

cogs.getcsv module

Module for save .csv from database

cogs.getcsv.get_pre_csv(userno: int)

save into .csv from sensor data and outdoor weather.

Parameters

userno (int) – User number from database

Returns

file name of saved .csv

Return type

fn1, fn2 (str)

cogs.refresh_class module

Module for predict user refresh.

class cogs.refresh_class.RFC

Bases: object

prepare data, build xgb model, predict data

build_model(userno: int, data: numpy.ndarray, target: numpy.ndarray)

build xgb model :param userno: User number from database :type userno: int :param data: Data to learn :type data: np.ndarray :param target: Target data :type target: np.ndarray

Returns: xgb

data_prep(df: pandas.core.frame.DataFrame, rf_list: list)

preprocessing dataframe

Parameters
  • df (pd.DataFrame) – Dataframe for preprocess

  • rf_list (list) – list of user refresh datetime

Returns

return train test set

Return type

data, target

pred_data(pdata: numpy.array, model=None)
pred_data_prep(df: pandas.core.frame.DataFrame)

preprossesing dataframe for model predict

Parameters

df (pd.DataFrame) – raw dataframe

Returns

data (pd.DataFrame)

cogs.refresh_class.build_model_test()

Test code for build model

cogs.refresh_class.class_pred_test()

Test code for model predict

cogs.spred module

Module for predict sensor data.

cogs.spred.sensor_predict(df: Optional[pandas.core.frame.DataFrame] = None, userno: int = 18)

predict future sensor data from prophet

Parameters
  • df (pd.DataFrame, optional) – Raw data for prediction. Defaults to None.

  • userno (int, optional) – User number from database. Defaults to 18.

Returns

Result Dataframe

Return type

fdf(pd.DataFrame)

cogs.spred.test()

Test code for plotting dataframe

cogs.sql module

Module for manage SQL query

class cogs.sql.SQL

Bases: object

connect to your DB and execute que, more

you must insert dbhost, dbuser, dbpw

after query, you need to close connection.

connclose()

close the session

insert_sql(table, nts, values)
Parameters
  • table (str) – insert table

  • nts (datetime) – primary datetime object

  • values (str) – sensor data list

Returns

sql result

Return type

rows (tuple)

insert_user_xy(table: str, id: int, nx: int, ny: int)

insert nx, ny into innovationdb.user

Parameters
  • table (str) – The name of the table to be referenced

  • id (int) – User number from database

  • nx (int) – x axis from user address

  • ny (int) – y axis from user address

Returns

sql result

Return type

rows(tuple)

view_table(table, userno: int, where, arg1: int, arg2)

custom select table query :param table: table to select :type table: str :param where: where arg to sql query :type where: str :param arg1: period to select :type arg1: int :param arg2: ‘day’ or ‘hour’ :type arg2: str

Returns

sql result

Return type

rows(tuple)

view_user_table(id: int)

select user table from database

Parameters

id (int) – User number from database

Returns

sql result

Return type

rows(tuple)

view_user_xy(id: int)

selecto nx, ny from user table

Parameters

id (int) – User number from database

Returns

sql result

Return type

rows(tuple)

Module contents

cog modules for innovation practice project