def run(V, in_data):
D = in_data
L, K = D.no_tasks, D.no_machines
C = D.tasks
E = []
F = [A for A in B(L)]
G = [0 for A in B(K)]
P = Q([A.weight for A in C])
F.sort(key=lambda idx: C[idx].weight / Q(C[idx].duration))
for W in B(L):
X = J(F)
H = []
for A in F:
if G[2] >= C[A].due_date: H.append(A)
if J(H) == 0: H = F
I = H[0]
for A in H[1:]:
if V.first_wins(I, A, C, P, G) == False: I = A
S = C[I]
E.append(I)
F.remove(I)
P -= S.weight
T = 0
for A in B(K):
G[A] = R(G[A], T) + S.duration[A]
T = G[A]
E = [A + 1 for A in E]
return U().s(D, M(score=O(E, C, K), schedule=N(D.no_tasks, E)))
def __getitem__(self, item):
"""
Get item
:param item:
:return:
"""
# Get target set
start_index = self.fold * self.k
test_set = self.indexes[start_index:start_index + self.fold_length]
train_set = np.remove(self.indexes, test_set)
# Train/test
if self.train:
return self.dataset[train_set[item]]
else:
return self.dataset[test_set[item]]
def predict(self, k, X):
#initiate the variables
num_test = X.shape[0]
Ypred = np.zeros(num_test, dtype=self.ytr.dtype)
min_indexes = np.array([])
for i in range(k):
for j in xrange(num_test):
distances = np.sum(np.abs(self.Xtr - X[i, :]), axis=1)
min_index = np.argmin(distances)
# add the minimum indexes
min_indexes = min_indexes.append(min_indexes, min_index)
distances = np.remove(distances, min_index)
# find the mode of the minimum indexes
min_index = stats.mode(min_indexes)
Ypred[i] = self.ytr[min_index]
return Ypred
def otherEdges(self, edge): return np.remove(self.edges, edge)
def otherEdges(self, edge):
return np.remove(self.edges, edge)
# augmentation
if augmentation:
print("Augmentation flipping")
X, Y = augmentation_flip(X, Y)
# preprocess
# am I messing up with the first dim here?
if preprocess:
print("Preprocessing: feature normalization")
X /= 255.0
X -= np.mean(X, axis=0)
if remove:
while i < X.shape[0]:
if Y[i] > limit or Y[i] < -limit:
Y = np.remove(Y, i)
X = np.remove(X, i)
else:
i+=1
# move channel axis
X = X.transpose(0, 3, 1, 2)
# subsample
totalSamples = X.shape[0]
num_train = int(totalSamples * (num_training_percentage / 100))
num_validation = int(totalSamples * (num_validation_percentage / 100))
mask = range(num_train)
X_train = X[mask]
y_train = Y[mask]
def _read_train_file(filename):
data=pd.read_csv(filename).as_matrix()
for i in range(data.shape[1]-2):
data=np.remove(data,1,0)
print("example data:",data[0])
return data
#3. sorting data data5=np.array([100,75,30,120,130,90]) np.sort(data5) #class challenge question # how to get descending order sorting ? -np.sort(-data5) #4. how to add new value # add new value 110 to data5 np.append(data5,110) #5. how to remove value from array # remove 30 from data5 np.remove(data5[2]) # syntax error, reason remove is part # of list function, not array function np.delete(data5,[2]) #6. replace value inside array # replace 75 to 80 data5[1]=80 print(data5) # ============================================================================= # Test date - 14th Aug 2019 # topics - till 12th Aug 2019 topics # 13th Aug 2019 - practice and doubt session # =============================================================================