由于有一个教程，我试图使用tensorflow，但是我真的在为必须使用它而苦苦挣扎。
目前，我训练了具有以下功能的模型：
def initialize_parameters(beta=0.05):
tf.set_random_seed(1)
#Regularization
if beta!=0:
regularizer = tf.contrib.layers.l2_regularizer(scale=beta)
else: regularizer=None
W1 = tf.get_variable('W1',[4,4,3,8],initializer=tf.contrib.layers.xavier_initializer(seed = 0), regularizer=regularizer)
W2 = tf.get_variable('W2',[2,2,8,16],initializer=tf.contrib.layers.xavier_initializer(seed = 0), regularizer=regularizer)
parameters = {"W1": W1,
"W2": W2}
return parameters, regularizer
def forward_propagation(X, parameters, regularizer=None):
# Retrieve the parameters from the dictionary "parameters"
W1 = parameters['W1']
W2 = parameters['W2']
# CONV2D: stride of 1, padding 'SAME'
Z1 = tf.nn.conv2d(X,W1,strides=[1,1,1,1],padding='SAME')
# RELU
A1 = tf.nn.relu(Z1)
# MAXPOOL: window 8x8, sride 8, padding 'SAME'
P1 = tf.nn.max_pool(A1, ksize=[1,8,8,1], strides=[1,8,8,1],padding='SAME')
# CONV2D: filters W2, stride 1, padding 'SAME'
Z2 = tf.nn.conv2d(P1,W2,strides=[1,1,1,1],padding='SAME')
# RELU
A2 = tf.nn.relu(Z2)
# MAXPOOL: window 4x4, stride 4, padding 'SAME'
P2 = tf.nn.max_pool(A2,ksize=[1,4,4,1],strides=[1,4,4,1],padding='SAME')
# FLATTEN
P2 = tf.contrib.layers.flatten(P2)
# FULLY-CONNECTED without non-linear activation function (not not call softmax).
# 6 neurons in output layer. Hint: one of the arguments should be "activation_fn=None"
Z3 = tf.contrib.layers.fully_connected(P2, num_outputs=6,activation_fn=None,weights_regularizer=regularizer)
return Z3
def compute_cost(Z3, Y, regularizer=None):
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits = Z3, labels = Y))
#Regularize
if regularizer is not None:
reg_variables = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
reg_term = tf.contrib.layers.apply_regularization(regularizer, reg_variables)
else:
reg_term = 0
cost += reg_term
return cost
之后，我用模型函数调用所有内容：
def model(X_train, Y_train, X_test, Y_test, learning_rate = 0.009,
num_epochs = 100, minibatch_size = 64, print_cost = True):
"""
Implements a three-layer ConvNet in Tensorflow:
CONV2D -> RELU -> MAXPOOL -> CONV2D -> RELU -> MAXPOOL -> FLATTEN -> FULLYCONNECTED
Arguments:
X_train -- training set, of shape (None, 64, 64, 3)
Y_train -- test set, of shape (None, n_y = 6)
X_test -- training set, of shape (None, 64, 64, 3)
Y_test -- test set, of shape (None, n_y = 6)
learning_rate -- learning rate of the optimization
num_epochs -- number of epochs of the optimization loop
minibatch_size -- size of a minibatch
print_cost -- True to print the cost every 100 epochs
Returns:
train_accuracy -- real number, accuracy on the train set (X_train)
test_accuracy -- real number, testing accuracy on the test set (X_test)
parameters -- parameters learnt by the model. They can then be used to predict.
"""
ops.reset_default_graph() # to be able to rerun the model without overwriting tf variables
tf.set_random_seed(1) # to keep results consistent (tensorflow seed)
seed = 3 # to keep results consistent (numpy seed)
(m, n_H0, n_W0, n_C0) = X_train.shape
n_y = Y_train.shape[1]
costs = [] # To keep track of the cost
# Create Placeholders of the correct shape
X, Y = tf.placeholder(dtype=tf.float32,shape=(None, n_H0, n_W0, n_C0),name="X"),tf.placeholder(dtype=tf.float32,shape=(None,6),name="Y")
# Initialize parameters
parameters = initialize_parameters()
# Forward propagation: Build the forward propagation in the tensorflow graph
Z3 = forward_propagation(X, parameters)
# Cost function: Add cost function to tensorflow graph
cost = compute_cost(Z3, Y)
# Backpropagation: Define the tensorflow optimizer. Use an AdamOptimizer that minimizes the cost.
optimizer = tf.train.AdamOptimizer(learning_rate).minimize(cost)
# Initialize all the variables globally
init = tf.global_variables_initializer()
# Start the session to compute the tensorflow graph
with tf.Session() as sess:
# Run the initialization
sess.run(init)
# Do the training loop
for epoch in range(num_epochs):
minibatch_cost = 0.
num_minibatches = int(m / minibatch_size) # number of minibatches of size minibatch_size in the train set
seed = seed + 1
minibatches = random_mini_batches(X_train, Y_train, minibatch_size, seed)
for minibatch in minibatches:
# Select a minibatch
(minibatch_X, minibatch_Y) = minibatch
# Run the session to execute the optimizer and the cost, the feedict should contain a minibatch for (X,Y).
_ , temp_cost = sess.run([optimizer, cost],feed_dict={X:minibatch_X,Y:minibatch_Y})
minibatch_cost += temp_cost / num_minibatches
# Print the cost every epoch
if print_cost == True and epoch % 5 == 0:
print ("Cost after epoch %i: %f" % (epoch, minibatch_cost))
if print_cost == True and epoch % 1 == 0:
costs.append(minibatch_cost)
# plot the cost
plt.plot(np.squeeze(costs))
plt.ylabel('cost')
plt.xlabel('iterations (per tens)')
plt.title("Learning rate =" + str(learning_rate))
plt.show()
# Calculate the correct predictions
predict_op = tf.argmax(Z3, 1)
correct_prediction = tf.equal(predict_op, tf.argmax(Y, 1))
# Calculate accuracy on the test set
accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float"))
print(accuracy)
train_accuracy = accuracy.eval({X: X_train, Y: Y_train})
test_accuracy = accuracy.eval({X: X_test, Y: Y_test})
print("Train Accuracy:", train_accuracy)
print("Test Accuracy:", test_accuracy)
return train_accuracy, test_accuracy, parameters
到目前为止，还不错，但是我无法根据新数据预测模型。 目前，我已经尝试过类似的事情：
X = scipy.misc.imresize（my_image，size =（64,64））。reshape（（1,64,64,3））/ 255。
with tf.Session() as sess:
x = tf.placeholder(tf.float32, shape=(None,64, 64, 3))
z3 = forward_propagation(x, parameters)
#soft = tf.nn.softmax(z3)
p = tf.argmax(z3,axis=1)
#init = tf.global_variables_initializer()
#sess.run(init)
prediction = sess.run(p, feed_dict = {x: X})
print(sess.run(z3, feed_dict = {x: X}))
print(prediction)
返回错误：尝试使用未初始化的值fully_connected_1 / biases [[Node：fully_connected_1 / biases / read = IdentityT = DT_FLOAT，_class = [“ loc：@ fully_connected_1 / biases”]，_ device =“ / job：localhost / replica：0 / task：0 / cpu：0”]] < / strong>
甚至：
def predict(X, parameters):
tf.reset_default_graph()
W1 = parameters["W1"]
W2 = parameters["W2"]
params = {"W1": W1,
"W2": W2}
x = tf.placeholder(tf.float32, shape=(None,64, 64, 3))
z3 = forward_propagation(x, params)
p = tf.argmax(z3)
sess = tf.Session()
prediction = sess.run(p, feed_dict = {x: X})
return prediction
但是当我用图像运行函数时，出现错误 Tensor（“ W1：0”，shape =（4，4，3，8），dtype = float32_ref）必须来自同一张图作为Tensor（“ Placeholder：0”，shape =（?, 64，64，3），dtype = float32）
如何使用训练有素的模型（我想我的参数存储在可变参数中）谢谢
答案 0 :(得分：0)
好吧，这可能不是最好的解决方案，但这就是我的做法：
首先，我保存该图并向集合中添加Forecast_op
def model(X_train, Y_train, X_test, Y_test, learning_rate = 0.003,
num_epochs = 1000, minibatch_size = 64, print_cost = True,beta = 0.05):
... some code that is actually the same than before ...
saver = tf.train.Saver()
tf.add_to_collection('predict_op', predict_op)
saver.save(sess, './my-test-model')
return train_accuracy, test_accuracy, parameters
然后加载
之前创建的图形## Predict the classification of the loaded image
## I am using the default data flow graph to run predictions
tf.reset_default_graph()
with tf.Session() as sess:
## Load the entire model previuosly saved in a checkpoint
print("Load the model from path", checkpoint_path)
the_Saver = tf.train.import_meta_graph(checkpoint_path + '.meta')
the_Saver.restore(sess, checkpoint_path)
## Identify the predictor of the Tensorflow graph
predict_op = tf.get_collection('predict_op')[0]
## Identify the restored Tensorflow graph
dataFlowGraph = tf.get_default_graph()
## Identify the input placeholder to feed the images into as defined in the model
x = dataFlowGraph.get_tensor_by_name("X:0")
## Predict the image category
prediction = sess.run(predict_op, feed_dict = {x: my_image_work})
print("\nThe predicted image class is:", str(np.squeeze(prediction)))
它似乎不是最有效的解决方案，但它有效！ 如果有更好的方法，请随时分享：）