args = parse_args()
np.random.seed(args.seed)
prob_args = {
"prob_name": args.prob_name,
"base_path": args.base_path,
"return_d3mds": True,
"use_schema": args.use_schema,
"strict": True,
}
t = time()
train, test = uv_load(args.base_path, args.prob_name)
import numpy as np
from sklearn.decomposition import PCA
def tss_to_numpy(bag=''):
keys = list(bag.keys())
remove = ['Type', 'Samples', 'Size', 'Labels']
[keys.remove(r) for r in remove]
return np.vstack([bag[i].data for i in keys])
from src.timeseries.TimeSeriesLoader import uv_load
from src.transformation.SFA import *
symbols = 8
wordLength = 16
normMean = False
def sfaToWord(word):
word_string = ""
alphabet = "abcdefghijklmnopqrstuv"
for w in word:
word_string += alphabet[w]
return word_string
train, test = uv_load("Gun_Point")
sfa = SFA("EQUI_DEPTH")
sfa.fitTransform(train, wordLength, symbols, normMean)
sfa.printBins()
for i in range(test["Samples"]):
wordList = sfa.transform2(test[i].data, "null")
print(str(i) + "-th transformed time series SFA word " + "\t" + sfaToWord(wordList))
def __init__(self, ucr_data):
self.sfa = {}
self.train, self.test = uv_load(ucr_data) # store train data only
from src.classification.WEASELClassifier import *
from src.classification.BOSSEnsembleClassifier import *
from src.classification.BOSSVSClassifier import *
from src.classification.ShotgunEnsembleClassifier import *
from src.classification.ShotgunClassifier import *
Datasets = [ #"Coffee",
# "Beef",
# "ECG200",
"Gun_Point",
# "BeetleFly"
]
for data in Datasets:
train, test = uv_load(data)
#The WEASEL Classifier
weasel = WEASELClassifier(data)
scoreWEASEL = weasel.eval(train, test)[0]
print(data + "; " + scoreWEASEL)
#The BOSS Ensemble Classifier
boss = BOSSEnsembleClassifier(data)
scoreBOSS = boss.eval(train, test)[0]
print(data + "; " + scoreBOSS)
#The BOSS VS Classifier
bossVS = BOSSVSClassifier(data)
scoreBOSSVS = bossVS.eval(train, test)[0]
print(data + "; " + scoreBOSSVS)
from src.timeseries.TimeSeriesLoader import uv_load
from src.timeseries.TimeSeriesLoader import mv_load
from src.utils import logger
import src.utils.parameters as params
FIXED_PARAMETERS = params.load_parameters()
logpath = FIXED_PARAMETERS["log_path"] + FIXED_PARAMETERS[
'test'] + "_" + FIXED_PARAMETERS['dataset'] + ".log"
logger = logger.Logger(logpath)
logger.Log("FIXED_PARAMETERS\n %s" % FIXED_PARAMETERS)
try:
train, test = uv_load(FIXED_PARAMETERS['dataset'], logger=logger)
except:
train, test = mv_load(FIXED_PARAMETERS['dataset'],
useDerivatives=True,
logger=logger)
try:
##=========================================================================================
## Multivariate Classifier Tests
##=========================================================================================
if FIXED_PARAMETERS['test'] == 'MUSE':
logger.Log("Test: MUSE")
from src.classification.MUSEClassifier import *
muse = MUSEClassifier(FIXED_PARAMETERS, logger)
scoreMUSE = muse.eval(train, test)[0]
logger.Log("%s: %s" % (FIXED_PARAMETERS['dataset'], scoreMUSE))
##=========================================================================================
## Univariate Classifier Tests