def __init__(self, eps, minPts, filePath):
self.filePath = filePath
self.minPts = minPts
self.label = 0
self.pixelLabels = collections.defaultdict(list)
self.visitedPixels = []
self.mongoConnectInstance = mongoConnect.MongoDBConnector(
"QuickDBScanDB")
def __init__(self, filePath):
img = cv2.imread(filePath, cv2.IMREAD_COLOR)
self.pixelList = []
for i in range(len(img)):
for j in range(len(img[0])):
self.pixelList.append([
i, j, img[i][j]
]) #position x, position y, rgb code (numpy array)
self.mongoConnectInstance = mongoConnect.MongoDBConnector(
"QuickDBScanDB")
def __init__(self, eps, dataset):
self.eps = eps
self.dataset = dataset
#clean collection
#number of dimension
self.no_dims = 2
self.mongoConnectInstance = mongoConnect.MongoDBConnector(
"QuickDBScanDB")
self.mongoConnectInstance.dropCollection("kdTreeDBSCAN")
def __init__(self, eps, pivotChoosingStrategy):
#int
self.eps = eps
#number of dimension
self.no_dims = 2
#pivot choosing strategy: 1 = corner, 2 = random
self.pivotChoosingStrategy = pivotChoosingStrategy
self.mongoConnectInstance = mongoConnect.MongoDBConnector(
"QuickDBScanDB")
#clean collection
self.mongoConnectInstance.dropCollection("quickDBSCAN")
def createEpsChains(self):
self.mongoConnectInstance = mongoConnect.MongoDBConnector(
"QuickDBScanDB")
results = self.kdIndex.query_ball_tree(self.kdIndex, self.eps)
for idx in range(len(self.dataset)):
for result in self.dataset[results[idx]]:
self.upsertPixelValue('kdTreeDBSCAN', {
"$or": [{
"bucket": []
}, {
"bucket": [result[0], result[1]]
}]
}, [[result[0], result[1]],
[self.dataset[idx][0], self.dataset[idx][1]]], False)
self.upsertPixelValue(
'kdTreeDBSCAN', {
"$or": [{
"bucket": []
}, {
"bucket":
[self.dataset[idx][0], self.dataset[idx][1]]
}]
}, [[result[0], result[1]],
[self.dataset[idx][0], self.dataset[idx][1]]])
import numpy as np
from gensim.models.doc2vec import Doc2Vec, TaggedDocument
from nltk.corpus import stopwords
import mongoConnect
from nltk.tokenize import RegexpTokenizer
mongoDb = mongoConnect.MongoDBConnector('ARXIVRobert6Clusters')
stemmers = ["porterStemmedWords", "lancasterStemmedWords", "lemmatizedWords"]
stop_words = set(stopwords.words('english'))
for stemmer in stemmers:
arxivRecordsCursor = mongoDb.getRecords("documents", {}, {"_id":1, stemmer: 1, "category": 1})
docs = []
for document in arxivRecordsCursor:
documentWithoutStop = [w for w in document[stemmer] if not w in stop_words]
docs.append(documentWithoutStop)
taggedDocuments = [TaggedDocument(doc, [i]) for i, doc in enumerate(docs)]
model = Doc2Vec(vector_size=16, window=3, min_count=2, workers=4, epochs=200)
model.build_vocab(taggedDocuments)
model.train(taggedDocuments, total_examples=len(taggedDocuments), epochs=200)
X = []
def __init__(self, eps):
#int
self.eps = eps
self.mongoConnectInstance = mongoConnect.MongoDBConnector(
"QuickDBScanDB")
self.allPairs = []