def L_layer_model(X,
Y,
layers_dims,
learning_rate=0.5,
num_iterations=10000,
print_cost=False): #lr was 0.009
np.random.seed(1)
costs = []
parameters = functions.initialize_parameters_deep(layers_dims)
for i in range(0, num_iterations):
AL, caches = functions.L_model_forward(X, parameters)
cost = functions.compute_cost(AL, Y)
grads = functions.L_model_backward(AL, Y, caches)
parameters = functions.update_parameters(parameters, grads,
learning_rate)
if print_cost and i % 100 == 0:
print("Cost after iteration %i: %f" % (i, cost))
if print_cost and i % 100 == 0:
costs.append(cost)
#plt.ylabel('cost')
#plt.xlabel('iterations (per tens)')
#plt.title("Learning rate =" + str(learning_rate))
#plt.show()
return parameters
def L_layer_model(
X, y, layers_dims, learning_rate=0.01, num_iterations=3000,
print_cost=True, hidden_layers_activation_fn="relu"):
random.seed(version =2)
np.random.seed(random.randint(0,1000))
parameters = fn.initialize_parameters(layers_dims)
for i in range(num_iterations):
AL, caches = fn.L_model_forward(
X, parameters, hidden_layers_activation_fn)
cost = fn.compute_cost(AL, y)
grads = fn.L_model_backward(AL, y, caches, hidden_layers_activation_fn)
parameters = fn.update_parameters(parameters, grads, learning_rate)
if (i + 1) % 100 == 0 and print_cost:
print(f"The cost after {i + 1} iterations is: {cost:.4f}")
return parameters
def accuracy(X, parameters, y, activation_fn="relu"):
probs, caches = fn.L_model_forward(X, parameters, activation_fn)
probs = probs.T
for i in range(np.shape(probs)[0]):
flag = True
li = np.asarray(probs[i])
Max = max(li)
for j in range(np.shape(probs)[1]):
if probs[i][j]==Max and flag:
probs[i][j] = 1
flag = False
else:
probs[i][j] = 0
y = y.T
count = 0
acc = {}
for i in range(10):
acc[i] = 0
for i in range(np.shape(probs)[0]):
flag = True
for j in range(np.shape(probs)[1]):
if probs[i][j] != y[i][j]:
flag = False
break
elif probs[i][j] == 1:
acc[j] +=1
break
if flag:
count +=1
for i in range(10):
acc[i] *= 100/(np.shape(probs)[0]/np.shape(probs)[1])
accuracy = count * 100/(np.shape(probs)[0])
print(acc)
print()
return f"The accuracy rate is: {accuracy:.2f}%."
print()
#Подсчёт ответа на конкретном файле
print("Доступные файлы:")
showFiles()
print()
path = input("Введите название файла в папке working, включая расширение, введите \"exit\" для завершения : ")
print()
dirName = os.path.join(os.path.realpath(r"..\..\..\ "),"samples","working")
while(str(path)!="exit"):
path = os.path.join(dirName,path)
try:
X_test = generateTrain(1,path)
if mode == "0":
probs, caches = fn.L_model_forward(X_test, fn.outputWB(layers_dims,"best"), "relu")
else:
probs, caches = fn.L_model_forward(X_test, fn.outputWB(layers_dims), "relu")
print("answer:")
ans = np.asarray(probs.T[0])
Max = max(ans)
for i in range(len(ans)):
if np.all(ans[i] == Max):
print(i)
break;
except IOError:
print("Can't open image")
print()
print("Доступные файлы:")
showFiles()
print()