csvfile))
elif outputfile_file_type == 'arff':
df = pd.read_csv(csvfile, engine='python')
filename, ext = os.path.splitext(args.i)
if df['class'].dtype == np.float:
df['class'] = df['class'].astype(int)
temp = df['class']
df = df.drop(columns=['class'], axis=1)
df['class'] = temp
DimensionReduction_filename = os.path.abspath(
'./Results') + os.sep + args.c
pandas2arff.pandas2arff(df,
filename=r'./Results/{}'.format(args.c),
wekaname=filename,
cleanstringdata=False,
cleannan=True)
logger.info(
'Reduced dimensional dataset has been saved in the {}.'.format(
DimensionReduction_filename))
#clean_csv(csvfile)
elif outputfile_file_type == 'libsvm':
df = pd.read_csv(csvfile, engine='python')
for x in df.columns:
if x.lower() == 'class':
label = x
break
y = df[label]
def main(args_i,
args_s=1,
args_e=-1,
args_l=1,
args_m=-1,
args_t='f1',
args_c='',
args_o_=''):
logger = logging.getLogger()
logger.setLevel(logging.INFO)
log_path = os.getcwd() + os.sep + 'Logs' + os.sep
rq = time.strftime('%Y%m%d%H%M', time.localtime(time.time()))
log_name = log_path + rq + '.log'
logfile = log_name
fh = logging.FileHandler(logfile, mode='w')
fh.setLevel(logging.INFO)
ch = logging.StreamHandler()
ch.setLevel(logging.INFO)
# logging.basicConfig(level=logging.INFO,
# format='[%(asctime)s]: %(message)s') # logging.basicConfig函数对日志的输出格式及方式做相关配置
formatter = logging.Formatter('[%(asctime)s]: %(message)s')
# 文件
fh.setFormatter(formatter)
logger.addHandler(fh)
# 控制台
ch.setFormatter(formatter)
logger.addHandler(ch)
logger.info("---mrmd 2.0 start----")
args = parse_args()
args.i, args.s,args.e,args.l,args.m,args.t,args.c,args.o = \
args_i, args_s , args_e , args_l , args_m , args_t , args_c , args_o_
file = args.i
file_type = file.split('.')[-1]
if file_type == 'csv':
pass
elif file_type == 'arff':
file = arff2csv(file)
elif file_type == 'libsvm':
file = libsvm2csv(file)
else:
assert "format error"
# format : arff or libsvm to csv
plt.figure(figsize=(2 * 4.7, 1 * 4.7))
plt.subplot(1, 2, 1)
tsne_scatter(file)
if int(args.e) == -1:
args.e = len(pd.read_csv(file, engine='python').columns) - 1
global args_o
if args.o == None:
args.o = ''.join(args.i.split('.')[:-1]) + '.metrics.csv'
args_o = args.o
d = Dim_Rd(file, logger)
d.run(inputfile=file)
outputfile = os.getcwd() + os.sep + 'Results' + os.sep + os.path.basename(
args_o)
csvfile = os.getcwd() + os.sep + 'Results' + os.sep + os.path.basename(
args.c)
logger.info(
"The output by the terminal's log has been saved in the {}.".format(
logfile))
logger.info('metrics have been saved in the {}.'.format(outputfile))
plt.subplot(1, 2, 2)
tsne_scatter(csvfile)
pngpath = os.path.abspath('./Results') + os.sep + os.path.basename(
args.i) + '.png'
plt.savefig(pngpath)
logger.info(
'Scatter charts visualized by t-SNE dataset has been saved in the {}.'.
format(pngpath))
# 处理输出文件的类型
outputfile_file_type = args.c.split('.')[-1]
if outputfile_file_type == 'csv':
logger.info(
'Reduced dimensional dataset has been saved in the {}.'.format(
csvfile))
elif outputfile_file_type == 'arff':
df = pd.read_csv(csvfile, engine='python')
filename, ext = os.path.splitext(args.i)
if df['class'].dtype == np.float:
df['class'] = df['class'].astype(int)
temp = df['class']
df = df.drop(columns=['class'], axis=1)
df['class'] = temp
DimensionReduction_filename = os.path.abspath(
'./Results') + os.sep + args.c
pandas2arff.pandas2arff(df,
filename=r'./Results/{}'.format(args.c),
wekaname=filename,
cleanstringdata=False,
cleannan=True)
logger.info(
'Reduced dimensional dataset has been saved in the {}.'.format(
DimensionReduction_filename))
# clean_csv(csvfile)
elif outputfile_file_type == 'libsvm':
df = pd.read_csv(csvfile, engine='python')
for x in df.columns:
if x.lower() == 'class':
label = x
break
y = df[label]
X = df.drop(columns=label, axis=1)
inputfile = args.i
# filename ,ext = os.path.splitext(inputfile)
DimensionReduction_filename = os.path.abspath(
'./Results') + os.sep + args.c
dump_svmlight_file(X,
y,
DimensionReduction_filename,
zero_based=True,
multilabel=False)
# clean_tmpfile.clean_csv(csvfile)
logger.info(
'Reduced dimensional dataset has been saved in the {}.'.format(
DimensionReduction_filename))
else:
logger.info(
'Reduced dimensional dataset has been saved in the {}.'.format(
csvfile))
logger.info("---mrmd 2.0 end---")