for i in range(0, len(query)):
normQuery.append((query[i] - queryMean) / queryStd)
cumLB.append(0)
# Define best-so-far, as we are returning query distance < bsf
# It is k-nearest neighor seacrch, so bsf is initially set to INF.
bsf = float("inf")
k = 5
scBand = 5
countKim = 0
print('query mean =', queryMean)
print('query std =', queryStd)
print('query len =', len(query))
print('best-so-far =', bsf)
print('kNN neighbor =', k)
_, sortingOrder = sort.bubbleSort(normQuery)
print('Sakoe-Chiba =', scBand)
print('ordering =', sortingOrder)
plt.plot(normQuery)
plt.title('Normalized Query')
plt.show()
### Step 1: Read all raw data ###
# For fair comparision, ALL data are loaded in memory.
print('......', time.ctime(), ' start loading data ......')
script_dir = os.path.dirname(__file__)
read_path = os.path.join(script_dir, 'rawData/data.json')
textfile = open(read_path, "r")
# lowResNormQuery = [
# [max of block],
# [min of block]
# ]
normQuery, queryMean, queryStd = normalization.forQuery(query, n)
lowResNormQuery = normalization.forLowResQuery(normQuery, lowResLen, n)
# Define best-so-far, as we are returning query distance < bsf
# It is k-nearest neighor search, so bsf is initially set to INF.
bsf = float("inf")
print('query mean =', queryMean)
print('query std =', queryStd)
print('query len =', len(query))
print('best-so-far =', bsf)
print('kNN neighbor =', k)
_, sortingOrder = sort.bubbleSort(normQuery)
lowResNormQueryAbs = []
for i in range(0, len(lowResNormQuery[0])):
lowResNormQueryAbs.append(
min(abs(lowResNormQuery[0][i]), abs(lowResNormQuery[1][i])))
_, sortingOrderNorm = sort.bubbleSort(lowResNormQueryAbs)
# print('ordering for ED =', sortingOrder)
# print('ordering for LB_LowResED =', sortingOrderNorm)
# Run ED
input('Please Enter to run similarity search by UCR_ED')
print('......', time.ctime(), ' start running UCR_ED ......')
timeStart = time.time()