def __init__(self, fileinstructions, study_info={}, logfile=None):
'''Initialization of the stats computation.'''
describ_info = parse_instructions_file(fileinstructions)
self.fileinstructions = fileinstructions
self.info = describ_info
self.stats = None
self.study_info = study_info
self.logfile = Logger(logfile)
def __init__(self, logfile, pathfolder, precomputers=None, num_cores=None):
self._initialization()
self._initialization_spec()
self.pathfolder = pathfolder
self.pathdata = os.path.join(pathfolder, 'Cleaned/Results')
if num_cores is None:
self.num_cores = multiprocessing.cpu_count()
elif num_cores == 0:
self.num_cores = 1
else:
self.num_cores = num_cores
self.logfile = Logger(logfile) if type(logfile) == str else logfile
if precomputers is not None:
self.precomputers = precomputers
else:
self.precomputers = PrecomputerCollection(logfile,
pathfolder,
old_computed=True)
class Statistics():
"""The object which performs the computation of the statistics.
TODO
----
Check if the variables in the info are in the dataframe and act in
consequence.
create plots function to create from stats the plots.
- Transform this class as a Processer
"""
def __init__(self, fileinstructions, study_info={}, logfile=None):
'''Initialization of the stats computation.'''
describ_info = parse_instructions_file(fileinstructions)
self.fileinstructions = fileinstructions
self.info = describ_info
self.stats = None
self.study_info = study_info
self.logfile = Logger(logfile)
def compute_stats(self, dataframe, info=None):
'''Function to compute the statistics for all the columns.'''
## 0. Prepare inputs
self.info = self.info if info is None else info
# Tracking process with logfile
t00 = time.time()
self.logfile.write_log(message0 % self.fileinstructions)
self.logfile.write_log(message1)
## 1. Compute stats
stats = []
for i in self.info.index:
info_var = dict(self.info.iloc[i])
# Tracking process with logfile
t0 = time.time()
self.logfile.write_log(message1a % info_var['variables'])
# Computing stats of the i-th variable
stats.append(compute_stats(dataframe, info_var))
# Stop tracking the process
self.logfile.write_log(message2 % (time.time()-t0))
## 2. Save and return
self.stats = stats
# TODO: Order by column order!!!
countsnull = np.sum(dataframe.notnull())
aux = pd.DataFrame([countsnull, dataframe.shape[0]-countsnull],
columns=['non-null', 'null'])
self.study_info['global_stats'] = aux
# Stop tracking the process
self.logfile.write_log(message3 % (time.time()-t00))
self.logfile.write_log(message_close)
return stats
def to_latex(self, filepath=None):
## Tracking the process
t0 = time.time()
self.logfile.write_log(message1b)
## 1. Compute transformation
doc = describe2latex(self.study_info, self.stats)
## 2. Write output
if filepath is None:
return doc
else:
#Write doc
with open(filepath, 'w') as myfile:
myfile.write(doc)
myfile.close()
# Stop tracking the process
self.logfile.write_log(message2 % (time.time()-t0))
self.logfile.write_log(message_close)
def clean_stats(self, stats=None):
if stats is None:
stats = self.stats
stats = clean_dict_stats(stats)
return stats
""" Script to test cleanning task """ from Mscthesis.Cleanning import CleanProcess from pythonUtils.Logger import Logger inpath = '/home/tono/mscthesis/code/Data/pruebas_raw/raw1' outpath = '/home/tono/mscthesis/code/Data/pruebas_clean2' logfile = '/home/tono/mscthesis/code/Data/Outputs/Logs/log_clean.log' logger = Logger(logfile) cleaner = CleanProcess(logger) cleaner.clean(inpath, outpath)
def __init__(self, logfile, pathdata):
self._initialization()
self.logfile = Logger(logfile) if type(logfile) == str else logfile
self.pathdata = pathdata
def __init__(self, logfile, bool_inform=False):
"Instantiation of the class remembering it is a subclass of Processer."
self._initialization()
self.logfile = Logger(logfile) if type(logfile) == str else logfile
