exellentPractice.append(extract[0])
weakPractice.append(extract[1])
# excellentList = exellentPractice[w]
# weakList = weakPractice[w]
# excellentList = labsheetActiveWeeks[w]
# weakList = labsheetLessActiveWeeks[w]
ex1_excellent = graphLearning.mapNewLabel(ex1_excellent, reLabelIndex)
ex1_weak = graphLearning.mapNewLabel(ex1_weak, reLabelIndex)
ex2_excellent = graphLearning.mapNewLabel(ex2_excellent, reLabelIndex)
ex2_weak = graphLearning.mapNewLabel(ex2_weak, reLabelIndex)
excellentList = ex2_excellent.index
weakList = ex2_weak.index
graphLearning.visualiseMSTGraph(graph_all_weeks[11], excellentList, weakList,
reLabelIndex)
#----------------------------------------------
#Node embedding analysis
#----------------------------------------------
node_embeddings_weeks = []
for w in range(0, 12):
print('Week ' + str(w) + '...')
node2vec = Node2Vec(graph_all_weeks[w].graph,
dimensions=64,
walk_length=8,
num_walks=15,
p=0.1,
q=1)
model = node2vec.fit(window=8, min_count=1)
~ca1162019_transitionDataMatrixWeeks[w].index.isin(ex3_excellent_2019.index
)].index
student2020_excellent = ca1162020_transitionDataMatrixWeeks[w].loc[
ca1162020_transitionDataMatrixWeeks[w].index.isin(
ex3_excellent_2020.index)].index
student2020_weak = ca1162020_transitionDataMatrixWeeks[w].loc[
~ca1162020_transitionDataMatrixWeeks[w].index.isin(ex3_excellent_2020.index
)].index
studentCohort = {
"2018 2019 excellent": student2018_excellent.union(student2019_excellent),
"2018 2019 weak": student2018_weak.union(student2019_weak),
"2020 excellent": student2020_excellent,
"2020 weak": student2020_weak
}
graphLearning.visualiseMSTGraph(graph_all_weeks[w], studentCohort,
reLabelIndex)
import matplotlib.cm as cm
G = graph_all_weeks[9].graph
node_color = []
nodelist = []
nodelist = []
for n in G.nodes:
nodelist.append(n)
if n in student2018_excellent.union(student2018_weak):
node_color.append('blue')
elif n in student2019_excellent.union(student2019_weak):
node_color.append('red')
elif n in student2020_excellent.union(student2020_weak):
