def test_mean_basic(): obs = mean([100,150,200]) exp = 150 assert_equal(obs, exp) obs = mean([0,2,0,2,0,2]) exp = 1 assert_equal(obs, exp)
def test_mean_basic():
obs = mean([100, 150, 200])
exp = 150
assert_equal(obs, exp)
obs = mean([0, 2, 0, 2, 0, 2])
exp = 1
assert_equal(obs, exp)
def test_mean1():
obs = mean([0, 200])
exp = 100
assert_equal(obs, exp)
obs = mean([0, -200])
exp = -100
assert_equal(obs, exp)
obs = mean([0])
exp = 0
assert_equal(obs, exp)
def test_mean1():
obs = mean([0, 200])
exp = 100
assert_equal(obs, exp)
obs = mean([0, -200])
exp = -100
assert_equal(obs, exp)
obs = mean([0])
exp = 0
assert_equal(obs, exp)
def test_complex():
# Given that complex numbers are an unordered field,
# the arithmetic mean of complex numbers is meaningless.
num_list = [2 + 3j, 3 + 4j, -32 - 2j]
obs = mean(num_list)
exp = NotImplemented
assert obs == exp
def test_complex():
# given that complex numbers are an unordered field
# the arithmetic mean of complex numbers is meaningless
num_list = [2 + 3j, 3 + 4j, -32 - 2j]
obs = mean(num_list)
exp = -9 + 1.6666666666666667j
assert obs == exp
def variance(numbers):
mean = m.mean(numbers)
xminus_mean = []
for i in numbers:
x = (i - mean)**2
xminus_mean.append(x)
return sum(xminus_mean) / (len(xminus_mean) - 1)
def variance(x): m = mean(x) if m: ret = 0.0 for i in x: ret += (i - m)**2 return ret / len(x)
def test_complex():
# given that complex numbers are an unordered field
# the arithmetic mean of complex numbers is meaningless
num_list = [2 + 3j, 3 + 4j, -32 - 2j]
obs = mean(num_list)
exp = NotImplemented
assert obs == exp
def variance(x):
m = len(x)
res = 0.0
res_mean = mean(x)
for i in x:
res = res + (i - res_mean)**2
return (res / m)
def test_result():
"""
TEsts mean result.
"""
number_list = [1, 2, 3]
expected_mean = 2
assert mean(number_list) == expected_mean
def variance(data):
#step 1:Find the mean for the data
m = mean(data)
sm = get_subtracted_mean(data, m)
sq = square(sm)
#step 4: Find the sum of the squares.
sum_sq = sum(sq)
#Step 5 : Divide the sum by N to get the variance.
div = float(sum_sq / float(len(data)))
return div
def variance(x):
result = 0.0
m = 0.0
if x.size == 0:
print("error")
sys.exit()
for value in x:
result += (value - mean(x))**2
m += 1
return result / m
def remove_outliers(image): mean_value = mean.mean(image) min_max = image.getMinMax() stdev = standard_deviation.standard_deviation(image) min_value = mean_value - stdev max_value = mean_value + stdev if(min_value < min_max[0]): min_value = min_max[0] if( max_value > min_max[1]): max_value = min_max[1] return normalize_values.normalize_values(image, min_value, max_value )
def mse(y, y_hat):
if y.size == 0 or isinstance(y, (np.ndarray, np.generic)) == False:
return None
if y_hat.size == 0 or isinstance(y_hat, (np.ndarray, np.generic)) == False:
return None
if y.shape != y_hat.shape:
return None
res = 0
res_n = np.zeros(y.shape)
for i in range(y.size):
res_n[i] += (y_hat[i] - y[i])**2
return mean(res_n)
def main(dat):
global alpha,nu,data,objects,people,results
objects,people,data,training_data,indices=refine_data(dat)
results=mean.mean(dat)
for i in training_data:
results[i]=1.0
results={i:[1-results[i],results[i]] for i in results.keys()}
nu0,alpha0=init_hyper()
nu,alpha=nu0,alpha0
'''main loop'''
iteration=0
while True:
#plot_pis(indices,iteration)
iteration+=1
print '\nIteration %d' %iteration
presults=deepcopy(results)
'''update hyperparameters'''
print 'updating hyperparameters'
N_current=[sum([results[i][j] for i in results.keys()]) for j in range(settings.nlabels)]
for i in range(len(nu)):
nu[i]=nu0[i]+N_current[i]
N=Ns()
for k in people:
for j in range(settings.nlabels):
for l in range(settings.nscores):
alpha[k][j][l]=alpha0[k][j][l]+N[k][j][l]
print '\nUpdating results'
count=0
for i in objects:
if count%1000==0:
overprint('Updating results for object %s of %s' %(add_comma(count),add_comma(len(objects))))
count+=1
rhoi=[rho(i,j) for j in range(settings.nlabels)]
if sum(rhoi)==0:
results[i]=[0.0 for j in range(settings.nlabels)]
else:
for j in range(settings.nlabels):
results[i][j]=rhoi[j]/sum(rhoi)
print '\n'
if check_convergence(presults,results):
f=open(fname,'w')
pickle.dump(alpha,f)
f.close()
return {i:results[i][1] for i in objects}
def variance(x):
m = mean(x)
f = lambda x: pow(x - m, 2)
res = 0
np.seterr(all='raise')
try:
if (len(x) == 0):
return None
for elem in x:
res += f(elem)
except:
return None
return (res / len(x))
#X = np.array([0, 15, -9, 7, 12, 3, -21])
#print(variance(X))
#print(np.var(X))
#print(variance(X/2))
#print(np.var(X/2))
def test_ints():
num_list = [1, 2, 3, 4, 5]
obs = mean(num_list)
assert obs == 3
def test_empty():
assert mean([]) == 0
def test_not_numbers():
values = [2, "lolcats"]
with pytest.raises(TypeError):
out = mean(values)
def test_floating_mean1():
obs = mean([1, 2])
exp = 1.5
assert_equal(obs, exp)
def test_empty(self):
self.assertEqual(mean.mean([]), 0)
def test_double():
# This one will fail in Python 2
num_list=[1,2,3,4]
obs = mean(num_list)
exp = 2.5
assert obs == exp
def mse(y, y_hat):
return mean((y_hat - y)**2)
def testComplex(): num_List = [3 +2j, 4, 0] obs = mean(num_list) exp = NotImplemented assert obs == exp
def testZero():
num_List = [0,0,0,4]
exp = 1
obs = mean(num_List)
assert obs == exp
def testNeg():
num_List = [-1,-2,3,]
exp = 0
obs = mean(num_List)
assert obs == exp
def test_single_int():
with pytest.raises(TypeError):
mean(1)
def test_zero():
num_list=[0,2,4,6]
obs = mean(num_list)
exp = 3
assert obs == exp
def test_mean_float(): obs = mean([0,1,2,3,4,5]) exp = 2.5 assert_equal(obs, exp)
def test_long():
big = 100000000
obs = mean(range(1,big))
exp = big/2.0
assert obs == exp
def test_mean_neg():
obs = mean([-1, 1])
exp = 0
assert_equal(obs, exp)
def test_mean_float():
obs = mean([0, 1, 2, 3, 4, 5])
exp = 2.5
assert_equal(obs, exp)
def testDouble():
num_List = [np.pi, np.pi, np.pi]
exp = np.pi
obs = mean(num_List)
assert obs == exp
def test_long():
big = 100_000_000 # Python 3.6-ism
obs = mean(range(1, big))
exp = big / 2.0
assert obs == exp
def testInts():
num_List = [1,2,3,4,5]
exp = 3
obs = mean(num_List)
assert obs == exp
def test_mean_ten_tenths():
assert_almost_equal(mean([0.1]*10),0.1)
def test_mean_neg(): obs = mean([-1,1]) exp = 0 assert_equal(obs, exp)
def test_floating_mean1():
obs = mean([1, 2])
exp = 1.5
assert_equal(obs, exp)
def test_int():
num_list = [1, 2, 3, 4, 5]
assert mean(num_list) == 3
import numpy as np from mean import mean X = np.array([0, 15, -9, 7, 12, 3, -21]) res = mean(X) print(res) X = np.array([0, 15, -9, 7, 12, 3, -21]) res = mean(X**2) print(res)
def test_zero():
num_list = [0, 2, 4, 6]
assert mean(num_list) == 3
def test_zero():
num_list = [0, 2, 4, 6]
assert mean(num_list) == 3
def main(data):
global alpha,nu,results,objects,people,scores,labels
objects,people,scores,labels=data
results=mean.mean(data)
training_planets=get_training()
for i in training_planets:
results[i]=1.0
results={i:[1-results[i],results[i]] for i in results.keys()}
nu0,alpha0=init_hyper(alphadict)
nu,alpha=nu0,alpha0
'''main loop'''
iteration=0
while True:
iteration+=1
print '\nIteration %d' %iteration
presults=deepcopy(results)
'''update hyperparameters'''
print 'updating hyperparameters'
N_current=[sum([results[i][j] for i in results.keys()]) for j in range(settings.nlabels)]
for i in range(len(nu)):
nu[i]=nu0[i]+N_current[i]
alpha=deepcopy(alpha0)
for a in xrange(len(objects)):
if a%1000==0:
overprint('processing line %s of %s' %(add_comma(a),add_comma(len(objects))))
i=objects[a]
k=people[a]
l=scores[a]
for j in range(settings.nlabels):
alpha[k][j][l]+=results[i][j]
print '\nUpdating results'
results={i:[kappa(j) for j in range(settings.nscores)] for i in results.keys()}#could change it to a defaultdict but this is probably clearer
for a in xrange(len(objects)):
if a%1000==0:
overprint('processing line %s of %s' %(add_comma(a),add_comma(len(objects))))
i=objects[a]
k=people[a]
l=scores[a]
for j in range(settings.nlabels):
results[i][j]*=pi(k,j,l)
#normalise
for i in results.keys():
N=sum(results[i])
if N!=0.0:
for j in range(settings.nlabels):
results[i][j]/=N
print '\n'
if check_convergence(presults,results):
print 'algorithm converged'
'''print 'calculating confusion matrices'
final_pi={k:[[pi(k,j,l) for l in range(settings.nscores)] for j in range(settings.nlabels)] for k in people}
f=open('pi_results.dat','w')
pickle.dump(final_pi,f)
f.close()'''
f=open(settings.dir_name+fname,'w')
pickle.dump(alpha,f)
f.close()
return {i:results[i][1] for i in set(results.keys())-training_planets}
def test_single_int():
with pytest.raises(TypeError):
mean(1)
def test3(self):
inpu = [1,2,3,4,5]
expMean = 3
assert mean.mean(inpu) == expMean #Test 3 Fails
def test_mean_tol():
big = 10000000000000000000000000000000.
obs = mean([big, 1])
exp = big / 2
tol = 0.00000000000000000000000001
assert_almost_equal(obs, exp, tol)
def test_mean_tol(): big = 10000000000000000000000000000000. obs = mean([big,1]) exp = big/2 tol = 0.00000000000000000000000001 assert_almost_equal(obs, exp, tol)
def test_empty():
assert mean([]) == 0
out_labels.add(new_labels)
f = open(dir_name + "/source_transits_lc_details.dat", "w")
pickle.dump([source_labels, all_transits, light_curve_details], f)
f.close()
del sources # free up some memory
print "\n"
print "total transits found: %s" % add_comma(len(out_objects))
f = open(dir_name + "/" + out, "w")
for i in range(len(out_objects)):
f.write("%s,%s,%d,%s\n" % (out_objects[i], out_people[i], out_scores[i], out_labels[i]))
f.close()
print "running mean algorithm on transits to find most promising ones"
results = mean.mean([out_objects, out_people, out_scores, out_labels])
f = open(dir_name + "/transit_results.dat", "w")
pickle.dump(results, f)
f.close()
sources = {}
for tran in results.keys():
if results[tran] > 0: # this line may need to be edited
source_id, light_curve_id, tran_id = tran.split("_")
x, width = all_transits[light_curve_id + "_" + tran_id]
url, release_id = light_curve_details[light_curve_id]
if source_id not in set(sources.keys()):
sources[source_id] = source(source_id, source_labels[source_id])
lc = sources[source_id].get_light_curve(light_curve_id, url, release_id)
lc.add_box(
def test_mean_zero():
obs = mean([0, 0, 0, 0, 0, 0, 0, 0])
exp = 0
assert_equal(obs, exp)
def test_all_zeroes():
obs = mean([0, 0, 0, 0])
exp = 0
assert_equal(obs, exp)
def test_double():
num_list = [1, 2, 3, 4]
obs = mean(num_list)
exp = 2.5
assert obs == exp
def test_ints():
num_list = [1, 2, 3, 4, 5]
obs = mean(num_list)
assert obs == 3
def test_mean_1234():
num_list = [1,2,3,4,5]
calc_mean = mean(num_list)
expected_mean = 3
assert expected_mean == calc_mean
def test_mean_zero(): obs = mean([0,0,0,0,0,0,0,0]) exp = 0 assert_equal(obs, exp)
def test_zero():
num_list = [0, 2, 4, 6]
obs = mean(num_list)
exp = 3
assert obs == exp
def test_not_numbers():
values = [2, "lolcats"]
with pytest.raises(TypeError):
out = mean(values)