def __init__(self, header):
# header = [SampleNames,Column1,Column2,...]
self.header = header
self.table = []
self.nspaces = [len(h) for h in header]
self.num_temp = '%.6g'
parser.parse_args()
def GetOutDir(path, year, region, sub):
parser.parse_args()
if path is None: path = parser.args.basepath
if year is not None: path = os.path.join(path, 'Plots' + year)
if region is not None: path = os.path.join(path, region + 'Plots_EWK')
if sub is not None: path = os.path.join(path, sub)
if parser.args.sub is not None: path = os.path.join(path, parser.args.sub)
return path
def __init__(self):
self.opt = parser.parse_args()
self.transform = transforms.Compose([
transforms.RandomCrop(self.opt.imageSize * self.opt.upSampling),
transforms.ToTensor()
])
data_route = r'E:\DataSet\DOTA\part1\images'
self.data_list = self.listdir(data_route)
print('Read ' + str(len(self.data_list)) + ' images')
from data.CCF import CCFDataSet
from torch.utils.data import DataLoader
import torchvision.transforms as transforms
from Parser import parser
from model.models import Generator
import torch
from PIL import Image
import cv2
import numpy as np
device = torch.device("cuda:0")
opt = parser.parse_args()
transform = transforms.Compose([
transforms.RandomCrop(opt.imageSize * opt.upSampling),
transforms.ToTensor()
])
transform2 = transforms.Compose([transforms.ToTensor()])
normalize = transforms.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])
scale = transforms.Compose([
transforms.ToPILImage(),
transforms.Resize(opt.imageSize),
transforms.ToTensor(),
transforms.Normalize(mean=[0.5, 0.5, 0.5], std=[0.5, 0.5, 0.5])
])
unnormalize = transforms.Normalize(mean=[-1, -1, -1], std=[2, 2, 2])
generator = Generator(16, opt.upSampling)
from Parser import parser
import pickle
from os import path
import sys
#Checking for dictionary
if not path.exists('./VocabDictionary.pkl'):
with open('VocabDictionary.pkl', 'wb') as f:
pickle.dump({}, f, protocol=pickle.HIGHEST_PROTOCOL)
if __name__ == '__main__':
parser.parse_args(sys.argv[1:])
def __init__(self,
year=None,
region=None,
lumi=None,
path=None,
config=None,
autovar=None,
useMaxLumi=False,
show=True,
blinded=None):
parser.parse_args()
self.year = year
self.region = region
self.show = show
self.setPath(path)
self.setConfig(config)
if self.region is None: self.region = GetRegion()
self.setLumi(lumi, useMaxLumi)
self.autovar = autovar
if parser.args.autovar is not None: self.autovar = parser.args.autovar
if self.autovar is True: self.autovar = 0
self.isBlinded = blinded
if self.isBlinded is None:
self.isBlinded = parser.args.blinded
self.MCList = []
for mc in self.config.mclist:
self.MCList.append(mc)
self.SampleList = ["Data"] + self.MCList
self.processes = {}
datafile = DataFileMap[self.region]
if 'Ele' in self.region and self.year == '2017':
datalist = []
for type in ("SE", "SP"):
for era in sorted(self.lumimap.keys()):
datalist.append('%s_%s_%s' % (datafile, type, era))
else:
datalist = [
'%s_%s' % (datafile, era)
for era in sorted(self.lumimap.keys())
]
self.processes["Data"] = Process("Data",
datalist,
None,
'data',
year=self.year,
region=self.region)
for mc in self.MCList:
if self.region == "GammaCR" and mc == "QCD" and not parser.args.use_ga_qcd:
fakefiles = [
datafile.replace("Gamma", "QCDFake")
for datafile in datalist
]
self.processes[mc] = Process("QCDFake",
fakefiles,
None,
'bkg',
leg=self.config.legmap[mc],
color=self.config.colmap[mc],
year=self.year,
region=self.region)
continue
filelist = list(self.config.filemap[mc])
if mc in self.config.nlomap and not parser.args.no_nlo:
filelist += list(self.config.nlomap[mc])
self.processes[mc] = Process(mc,
filelist,
GetMCxsec(filelist, self.config.xsec),
'bkg',
leg=self.config.legmap[mc],
color=self.config.colmap[mc],
year=self.year,
region=self.region)
if self.region == "SignalRegion" and any(parser.args.signal):
self.setSignalInfo()
self.haddFiles()
if os.getcwd() != self.cwd: os.chdir(self.cwd)
def __init__(self,
year=None,
region=None,
lumi=None,
path=None,
config=None,
autovar=False,
useMaxLumi=False,
show=True):
parser.parse_args()
self.year = year
self.region = region
self.show = show
self.setPath(path)
self.setConfig(config)
if self.region is None: self.region = GetRegion()
self.setLumi(lumi, useMaxLumi)
self.autovar = autovar
if parser.args.autovar: self.autovar = True
self.isBlinded = False
self.MCList = []
for mc in self.config.mclist:
if parser.args.nlo and mc in self.config.nlomap:
self.MCList.append(self.config.nlomap[mc])
else:
self.MCList.append(mc)
self.SampleList = ["Data"] + self.MCList
self.processes = {}
datafile = DataFileMap[self.region]
if 'Ele' in self.region and self.year == '2017':
datalist = []
for type in ("SE", "SP"):
for era in sorted(self.lumimap.keys()):
datalist.append('%s_%s_%s' % (datafile, type, era))
else:
datalist = [
'%s_%s' % (datafile, era)
for era in sorted(self.lumimap.keys())
]
self.processes["Data"] = Process("Data",
datalist,
None,
'data',
year=self.year,
region=self.region)
for mc in self.MCList:
self.processes[mc] = Process(mc,
self.config.filemap[mc],
GetMCxsec(self.config.filemap[mc],
self.config.xsec),
'bkg',
leg=self.config.legmap[mc],
color=self.config.colmap[mc],
year=self.year,
region=self.region)
if self.region == "SignalRegion" and any(parser.args.signal):
self.setSignalInfo()
self.haddFiles()
if os.getcwd() != self.cwd: os.chdir(self.cwd)