'Fizyka statystyczna i termodynamika': 'Statistical Physics and Thermodynamics',
'Opracowanie danych doświadczalnych': 'Analysis of Experimental Data',
'Analiza matematyczna 2': 'Mathematical Analysis 2',
'Probabilistyka': 'Probability',
'Algebra z geometrią': 'Algebra and Geometry',
'Wstęp do fizyki jądrowej': 'Introduction to Nuclear Physics',
'Analiza matematyczna 3': 'Mathematical Analysis 3',
'Laboratorium fizyki 2': 'Physics Laboratory 2',
'Analiza matematyczna 1': 'Mathematical Analysis 1',
'Mechanika': 'Mechanics'
}
cl = 'L'
co_corr = np.corrcoef(win.getData(class_=cl), rowvar=0)
labels = [pl_en[x] for x in win.getCoursesNames()]
mds = MDS(n_components=2, dissimilarity='precomputed')
dists = np.empty((len(co_corr), len(co_corr)))
for ii in range(len(labels)):
for jj in range(len(labels)):
dists[ii][jj] = math.sqrt(2 * (1 - co_corr[ii][jj]))
pos = mds.fit(dists).embedding_
G = nx.Graph()
G.add_nodes_from(range(len(labels)))
textstr = ""
for ii, l in enumerate(labels):
textstr += str(ii) + " - " + l + "\n"
for jj in range(ii + 1, len(labels)):
d = dists[ii][jj]
'''spec: OE, MN, FK, FM''' # students = win.getStudents(shuffle=False) # cl = students[0].class_ # sp = students[0].spec # anno = [] # anno2 = [] # for ii, st in enumerate(students): # if st.class_ != cl: # anno.append((ii - 1, students[ii - 1].class_)) # cl = st.class_ # if st.spec != sp: # anno2.append((ii - 1, students[ii - 1].spec)) # sp = st.spec # anno.append(((len(students) - 1), students[-1].class_)) # anno2.append(((len(students) - 1), students[-1].spec)) # anno = anno + anno2 students = win.getData(shuffle=False, spec='FK') # win.shuffleIt(students, 2) # win.shuffleIt(students, 1) # st_corr = pairwise_distances(students, students, 'jaccard') co_corr = pairwise_distances(students.T, students.T, 'jaccard') heatmap.plotheat(co_corr, xlabels=win.getCoursesNames(), ylabels=win.getCoursesNames(), mode='special') # heatmap.plotheat(st_corr, changeTicks=False, annotation=anno, mode='special')
matplotlib.rc("font", **font)
matplotlib.rcParams["ps.useafm"] = True
matplotlib.rcParams["pdf.use14corefonts"] = True
matplotlib.rcParams["text.usetex"] = True
# win.shuffleIt(students, 2)
# win.shuffleIt(students, 1)
# st_corr = np.corrcoef(students)
# shuffle C
students = win.getData()
win.shuffleIt(students, 2)
co_corr = np.corrcoef(students, rowvar=0)
heatmap.plotheat(
co_corr,
xlabels=[pl_en[x][1] for x in win.getCoursesNames()],
ylabels=["" for x in win.getCoursesNames()],
rotation=90,
num_levels_=6,
mode="special",
vmin=vmin,
vmax=vmax,
showplot=False,
removebar=True,
)
savePic(name="c_heat_shufC.eps", path="/home/luke/Documents/APPA/img/fifth/")
# real
students = win.getData()
co_corr = np.corrcoef(students, rowvar=0)
heatmap.plotheat(
co_corr,
anno2 = []
for ii, st in enumerate(students):
if st.class_ != cl:
anno.append((ii - 1, students[ii - 1].class_))
cl = st.class_
if st.spec != sp:
anno2.append((ii - 1, students[ii - 1].spec))
sp = st.spec
anno.append(((len(students) - 1), students[-1].class_))
anno2.append(((len(students) - 1), students[-1].spec))
anno = anno + anno2
courses = win.getCourses()
'''correlation course-course'''
co_corr = np.empty([len(courses), len(courses)])
for ii, c1 in enumerate(courses):
for jj, c2 in enumerate(courses):
v1 = []
v2 = []
for st in students:
v1.append(st.grades[ii])
v2.append(st.grades[jj])
random.shuffle(v1)
random.shuffle(v2)
co_corr[ii, jj] = mutual_info_score(v1, v2)[0]
# print(len(students))
heatmap.plotheat(co_corr, xlabels=win.getCoursesNames(), ylabels=win.getCoursesNames())
heatmap.plotheat(st_corr, changeTicks=False, annotation=anno)
# heatmap.plotheat(st_corr, changeTicks=False)
'Podstawy elektroniki': 'Fundamentals of Electronics',
'Grafika inżynierska': 'Engineering Graphics',
'Metody matematyczne fizyki': 'Mathematical Methods of Physics',
'Elektronika w eksperymencie fizycznym': 'Electronics in Physical Experiment',
'Podstawy projektowania przyrządów wirtualnych': 'Fundamentals of Virtual Devices Design',
'Programowanie obiektowe': 'Object-Oriented Programming',
'Podstawy optyki': 'Fundamentals of Optics',
'Fizyka kwantowa': 'Quantum Physics',
'Fizyka statystyczna i termodynamika': 'Statistical Physics and Thermodynamics',
'Opracowanie danych doświadczalnych': 'Analysis of Experimental Data',
'Analiza matematyczna 2': 'Mathematical Analysis 2',
'Probabilistyka': 'Probability',
'Algebra z geometrią': 'Algebra and Geometry',
'Wstęp do fizyki jądrowej': 'Introduction to Nuclear Physics',
'Analiza matematyczna 3': 'Mathematical Analysis 3',
'Laboratorium fizyki 2': 'Physics Laboratory 2',
'Analiza matematyczna 1': 'Mathematical Analysis 1',
'Mechanika': 'Mechanics'
}
courses = win.getData().T
courses_names = win.getCoursesNames()
courses_names = [pl_en[x] for x in courses_names]
co_corr = np.corrcoef(win.getData(), rowvar=0)
for ii, c in enumerate(courses):
av = round(np.mean(c), 2)
var = round(np.var(c), 2)
cij = round(sum(co_corr[ii]) - 1.0, 2)
print(courses_names[ii], av, var, cij, sep=' & ')
'Fizyka kwantowa': ('Quantum Physics', 'QP'),
'Fizyka statystyczna i termodynamika': ('Statistical Physics and Thermodynamics', 'SPT'),
'Opracowanie danych doświadczalnych': ('Analysis of Experimental Data', 'AED'),
'Analiza matematyczna 2': ('Mathematical Analysis 2', 'MA2'),
'Probabilistyka': ('Probability', 'P'),
'Algebra z geometrią': ('Algebra and Geometry', 'AG'),
'Wstęp do fizyki jądrowej': ('Introduction to Nuclear Physics', 'INP'),
'Analiza matematyczna 3': ('Mathematical Analysis 3', 'MA3'),
'Laboratorium fizyki 2': ('Physics Laboratory 2', 'PL2'),
'Analiza matematyczna 1': ('Mathematical Analysis 1', 'MA1'),
'Mechanika': ('Mechanics', 'M')
}
co_corr = np.corrcoef(win.getData(class_='K'), rowvar=0)
labels = win.getCoursesNames()
G = nx.Graph()
G.add_nodes_from(range(len(labels)))
textstr = ""
for ii, l in enumerate(labels):
textstr += str(ii) + " - " + l + "\n"
for jj in range(ii + 1, len(labels)):
d = math.sqrt(2 * (1 - co_corr[ii][jj]))
G.add_edge(ii, jj, weight=d)
si = []
for n, nbrs in G.adjacency_iter():
w = 0
for nbr, eattr in nbrs.items():
w += 1 / eattr['weight']
'''course variance in function of its average'''
c_vars = []
c_avs = []
for c in data.T:
c_vars.append(np.var(c))
c_avs.append(np.mean(c))
lr.fit([[x] for x in c_avs], [[x] for x in c_vars])
pr = lr.predict([[x] for x in c_avs])
#
plt.plot(c_avs, c_vars, 'ro', markersize=10)
plt.plot(c_avs, pr, 'g-', label='fit', linewidth=3.0)
plt.xlim(3.2, 4.8)
plt.xlabel('mean value of course\'s grades')
plt.ylabel('variance of course\'s grades')
for ind, x, y in zip(range(len(win.getCoursesNames())), c_avs, c_vars):
plt.annotate(pl_en[win.getCoursesNames()[ind]][1], xy=(x, y), xytext=(-6, -15),
textcoords='offset points', fontsize=18)
# legend = "Legend:\n"
# legend += "\n".join([str(ii) + ': ' + pl_en[name]
# for ii, name in enumerate(win.getCoursesNames())])
# props = dict(boxstyle='round', facecolor='white', alpha=0.1)
# ax.text(0.65, 0.99, legend,
# transform=ax.transAxes,
# fontsize=8, verticalalignment='top', bbox=props)
plt.savefig('/home/luke/Documents/APPA/img/sixth/avs_var.svg', format='svg',
transparent=False, frameon=False, dpi=300, bbox_inches='tight')
#
sys.exit()
'''node force/strength with variances and avs (calculacated above) and pca'''
colourss = ['k*', 'go', 'm^', 'cv']
fc = []
sc = []
for ii, cl in enumerate(classes):
data = win.getData(class_=cl)
if cl is None:
cl = 'all'
avss = []
avsc = []
for s in data:
avss.append(np.mean(s))
for jj, c in enumerate(data.T):
m = np.mean(c)
avsc.append(m)
if m < 3.65 and m > 3.35 and cl == 'all':
fc.append(win.getCoursesNames()[jj])
if m < 4.3 and m > 4.01 and cl == 'all':
sc.append(win.getCoursesNames()[jj])
hist, bins = np.histogram(avss, bins=6, range=(3.0, 5.0))
bins = (bins[:-1] + bins[1:]) / 2
if cl == 'all':
plt.plot(bins, hist, colours[ii] + '-',
label='' + cl, linewidth=2.0, markersize=8)
else:
plt.plot(bins, hist, colours[ii] + '-', label='' + cl, markersize=8)
plt.xlabel('average grade')
plt.ylabel('number of students')