def disaggregate_chunk(self, mains,ep,n):
'''In-memory disaggregation.
Parameters
----------
mains : pd.Series of aggregate data
Returns
-------
appliance_powers : pd.DataFrame where each column represents a
disaggregated appliance. Column names are the integer index
into `self.model` for the appliance in question.
'''
up_limit = len(mains)
mains.fillna(0, inplace=True)
X_batch = np.array(mains)
#print(X_batch)
#X = laplace(X_batch,ep,n)
X_batch = np.reshape(X_batch, (X_batch.shape[0],1,1))
#print(X_batch.shape)
print(X_batch)
X_batch_1 = copy.deepcopy(X_batch)
X_batch = laplace_generator(X_batch_1,n,ep)
#print(X_batch)
pred = self.model.predict(X_batch, batch_size=128)
pred = np.reshape(pred, (len(pred)))
column = pd.Series(pred, index=mains.index[:len(X_batch)], name=0)
appliance_powers_dict = {}
appliance_powers_dict[0] = column
appliance_powers = pd.DataFrame(appliance_powers_dict)
return appliance_powers
def laplace(test_elec,ep):
num = test_elec.shape[0]
array = np.zeros(num)
for i in range(num):
array[i] = test_elec.iloc[i]
print(array)
n = 10
test_fix = laplace_generator(array,n,ep)
for i in range(num):
test_elec.iloc[i] = test_fix[i]
return test_elec
adapted_event = 0
adapted_errors = 0
multi_switches_count = 0
print()
ep = 0.009
for (fold, priors, testing) in folds:
del priors
tm_start = time()
sshmm = sshmms[fold]
obs_id = list(testing)[0]
obs1 = list(testing[obs_id])
obs2 = copy.deepcopy(obs1)
#print(obs1)
obs = laplace_generator(obs2, n, ep)
#print(obs)
hidden = [i for i in testing[labels].to_records(index=False)]
print()
print(
'Begin evaluation testing on observations, compare against ground truth...'
)
print()
pbar = ''
pbar_incro = len(testing) // 20
"""
noise
"""
for i in range(1, len(obs)):
multi_switches_count += (sum(
c = 0
#ep = 100-j*0.8
#ep = 100
ep = 0.001 * (j + 1)
k = 0
for i in range(1):
f1 = open(
"D:/project/CS5/CS8/CS/Viterbi - 副本/synctic_data02" +
str(sparsity) + ".txt", 'a')
f = open(
"D:/project/CS5/CS8/CS/Viterbi - 副本/synctic_data02_2" +
str(sparsity) + ".txt", 'a')
time_start = ti.time()
#print(y_fix_use)
#y_fix = y_fix_use
y_fix = laplace_generator(y_fix_use, n, ep)
#print(y_fix)
delta_x, delta_y, _size, _app = data_preprocess(
ground_truth, y_fix)
error = []
for j in range(delta_y.shape[0]):
error.append(
one_shot_compressed_sensing_decode(
delta_x[j], delta_y[j], p, delta))
acc = 1 - perform_matrix(error, _size, _app)
print("acc", acc)
temp = k
a += acc
c += 1
k = a / c
if temp - k < 0.00001 and i > 10: