def al_pkdd(dataset_number):
"""
Load Activation Log number `dataset_number`
Parameters
----------
dataset_number: int
index of the Activation Log, lower than 500
Returns
-------
al: logs.ActivationLog
The Activation Log with given index
Notes
-----
The Activation Logs are saved with 500 per file (to limit the number of files). This function handles the retrieval.
"""
assert dataset_number < 100
df = dataframe_operations \
.import_df(f'data/public/PKDD/ActivationLogs2.csv') \
.iloc[dataset_number * 1000: (dataset_number + 1) * 1000]
return logs.ActivationLog(df)
def al2dsx(dataset, hierarchy, additional_rp=None):
"""
Wrapper to adapt the dataset and the hierarchy to the new formulations of Subsection \\ref{sec:compopt:ws2w}
Parameters
----------
dataset: Counter, logs.ActivationLog, str, or Path.
The value of :math:`\mathcal{D'}`. If ActivationLog, str or Path, it is (imported and) transformed into a bag
hierarchy: hierarchy.Hierarchy
Hierarchy object defining the relations between (super-)categories
additional_rp: Counter
Additional r_prime values to be added apart from the dataset. These values are not part of visit_ids_per_rp
Returns
-------
dataset: DSX
D initiated with the adaption presented in Subsection \\ref{sec:compopt:ws2w}
- :math:`r'`, :math:`\mathcal{R}'`
- hierarchy class as given
- invoice_ids_per_rp, reference to original iids that belong to each rp
"""
# import the data
if isinstance(dataset, str) or isinstance(dataset, Path):
dataset = logs.ActivationLog(dataset)
else:
assert isinstance(dataset, logs.ActivationLog)
assert isinstance(hierarchy, Hierarchy)
if additional_rp is None:
additional_rp = Counter()
else:
assert isinstance(additional_rp, Counter)
df = dataset.df.copy()
h_vector = [
bitops.subset2int(hierarchy[k], hierarchy.c) for k in hierarchy.sc
]
def row2rp(row):
r = int(''.join([str(row[i]) for i in hierarchy.c]), 2)
return r + (int(
''.join([
str(1 if (hamming_weight(ci & r) > 0) else 0)
for ci in h_vector
]), 2) << hierarchy.num_cat)
df['rp'] = df.apply(row2rp, axis=1)
dataset = Counter(df['rp'])
dataset += additional_rp
# noinspection PyTypeChecker
visit_ids_per_rp = {
v: list(df[df['rp'] == v].index)
for v in df['rp'].unique()
}
return DSX(dataset, hierarchy, visit_ids_per_rp)
def al_apkdd(dataset_number):
assert dataset_number < 100
sr = pd.Series(
data=range(100000)).apply(lambda x: (x % 1000) // 10 == dataset_number)
df = dataframe_operations.import_df(
f'data/public/PKDD/ActivationLogs2.csv')[sr]
return logs.ActivationLog(df)
def al_data(dataset_exponent, dataset_number):
assert isinstance(dataset_exponent, int)
assert isinstance(dataset_number, int)
if dataset_exponent in [3, 4, 5]:
# These are saved with multiple values in one log
assert dataset_number < 10**(7 - dataset_exponent)
size = 10**dataset_exponent
number_of_datasets_per_file = 500000 // size
file_number = dataset_number // number_of_datasets_per_file
df = dataframe_operations.import_df(
retailer_folder / f'D{dataset_exponent}' / f'{file_number}.csv')
dataset_number_in_file = dataset_number % number_of_datasets_per_file
df = df.iloc[dataset_number_in_file *
size:(dataset_number_in_file + 1) * size]
return logs.ActivationLog(df)
elif dataset_exponent in [6, 7]:
assert (dataset_exponent == 6
and dataset_number < 10) or (dataset_exponent == 7
and dataset_number == 0)
return logs.ActivationLog(retailer_folder / f'D{dataset_exponent}' /
f'{dataset_number}.csv')
else:
raise ValueError('Illegal dataset_exponent')
def al_extended_example2():
return logs.ActivationLog(public_data_folder / 'samples' /
'implementation guide' / 'extended_example2.csv',
categories=example_cat_list())
def al_running_example():
return logs.ActivationLog(public_data_folder / 'samples' /
'implementation guide' / 'running_example.csv',
categories=example_cat_list())
def al_sample(exp):
assert exp in range(1, 5), 'given exp must be in 1...4'
return logs.ActivationLog(public_data_folder / 'samples' / 'real dataset' /
f'S{exp}.csv')