Question

我在tensorflow-gpu 1.8.0中实现了MLP，我使用Hyperopt来查找最佳参数配置。每当损失函数进一步最小化时，就会创建一个检查点文件。检查点文件总是被覆盖，在流程结束时，我只获得了这些文件：

检查点

Model_1_checkpoint.ckpt.data-00000-的-00001

Model_1_checkpoint.ckpt.index

Model_1_checkpoint.meta

我在代码下方显示训练MLP（在函数内）：

# Hyperparameters
n_step= np.round(parameters['step'],3)
n_hidden= np.int(parameters['number_neurons'])
n_bias= np.round(parameters['bias'],3)
n_batch= np.int(parameters['batch'])

# General variables
N_instances= xtrain_data_1_T60.shape[0]
N_input= xtrain_data_1_T60.shape[1]
N_classes= enc_ytrain_data_1_T60.shape[1]
N_epochs= 500
display_step= 100

# Reset graph
tf.reset_default_graph()

# Placeholders
X= tf.placeholder(name= "Logs", dtype= tf.float32, shape= [None, N_input])
y= tf.placeholder(name= "Facies", dtype= tf.float32, shape= [None, N_classes])

# MLP network architecture
input_layer= tf.layers.dense(X, units= N_input, activation= None, 
                             kernel_initializer= tf.keras.initializers.glorot_normal(1969),
                             bias_initializer= tf.keras.initializers.Zeros())

hidden_layer= tf.layers.dense(input_layer, units= n_hidden, activation= tf.nn.tanh, 
                              kernel_initializer= tf.keras.initializers.he_normal(1969),
                              bias_initializer= tf.keras.initializers.Constant(n_bias))

output_layer= tf.layers.dense(hidden_layer, units= N_classes, activation= tf.nn.softmax,
                              kernel_initializer= tf.keras.initializers.he_normal(1969),
                              bias_initializer= tf.keras.initializers.Zeros(), name= "mlp_output")

loss_op= tf.reduce_mean(tf.keras.backend.binary_crossentropy(y, output_layer))

optimizer= tf.train.GradientDescentOptimizer(learning_rate= n_step).minimize(loss_op)

# Initialize variables
init= tf.initialize_all_variables() #tf.global_variables_initializer()


with tf.Session() as sess:
    sess.run(init)

    # Training loop
    for epoch in range(0, N_epochs):
        avg_cost = 0.
        total_batch= np.int(N_instances/n_batch)
        start_idx= 0
        end_idx= n_batch

        for i in range(0, total_batch):
            batchx= xtrain_data_1_T60[start_idx:end_idx,:]
            batchy= enc_ytrain_data_1_T60[start_idx:end_idx,:]

            _, c= sess.run([optimizer, loss_op], feed_dict= {X: batchx, y: batchy})
            avg_cost += c/total_batch

            # Set next batch
            start_idx += n_batch
            end_idx += n_batch
            if (end_idx > N_instances):
                end_idx= N_instances

        if (epoch % display_step == 0):
            print("Epoch : %03d/%03d cost : %.4f\n"%(epoch, N_epochs, avg_cost))

    print("Optimization finished\n")

    prediction_1= sess.run(output_layer, feed_dict= {X: xvalidation_data_1_V40})
    prediction_1= prediction_1.argmax(axis= 1) + 1

    # Initialize a saver to save the current best model
    saver= tf.train.Saver(max_to_keep= 1)

    # Only check for prediction results with 3 lithofacies. Otherwise, I assign a dummy error and accuracy
    if len(np.unique(prediction_1)) == 3:
        error= 1. - metrics.recall_score(yvalidation_data_1_V40, prediction_1, average= 'micro')
        accuracy= metrics.accuracy_score(yvalidation_data_1_V40, prediction_1)

        global temp_error
        if (error < temp_error):
            temp_error= error
            saver.save(sess, '{}/{}'.format(checkpoint_path, checkpoint_name))
            print("Best model saved in file: ", '{}/{}'.format(checkpoint_path, checkpoint_name))
            print()
    else:
        error= 3
        accuracy= 0.00

    print("Error: {}".format(error))
    print("Accuracy: {:.2%}".format(accuracy))
    print("Predicted number of lithofacies: {}\n".format(len(np.unique(prediction_1))))

    sess.close()

然后，在同一个脚本中，我恢复保存检查点以计算预测：

tf.reset_default_graph()

# Restore the best model and predict again
with tf.Session() as sess:
    new_saver= tf.train.import_meta_graph(checkpoint_path + "/" + checkpoint_name + ".meta")
    new_saver.restore(sess, checkpoint_path)

    # Retrieve placeholder from restored graph
    X= best_model_1.get_tensor_by_name('Logs:0')

    # Retrieve output layer of MLP network to compute predictions
    pred= best_model_1.get_tensor_by_name('mlp_output/kernel:0')

    model_prob_density_1= sess.run(pred, feed_dict= {X: voting_data})

不幸的是，该行＆＃34; new_saver.restore（...）＆＃34;导致以下错误消息：

NotFoundError（参见上面的回溯）：检查点中找不到密钥密集/偏差 [[节点：save / RestoreV2 = RestoreV2 [dtypes = [DT_FLOAT，DT_FLOAT，DT_FLOAT，DT_FLOAT，DT_FLOAT，DT_FLOAT]，_ device =＆＃34; / job：localhost / replica：0 / task：0 / device：CPU：0＆＃34;]（_ arg_save / Const_0_0，save / RestoreV2 / tensor_names，save / RestoreV2 / shape_and_slices）]] [[节点：save / RestoreV2 / _9 = _Recvclient_terminated = false，recv_device =＆＃34; / job：localhost / replica：0 / task：0 / device：GPU：0＆＃34;，send_device =＆＃34; / job ：localhost / replica：0 / task：0 / device：CPU：0＆＃34;，send_device_incarnation = 1，tensor_name =＆＃34; edge_14_save / RestoreV2＆＃34;，tensor_type = DT_FLOAT，_device =＆＃34; / job：本地主机/复制：0 /任务：0 /设备：GPU：0＆＃34;]]

我回到代码的第一个块来尝试这些更改：

a）saver = tf.train.Saver（max_to_keep = 1） - ＆gt; saver = tf.train.Saver（tf.global_variables（），max_to_keep = 1）

b）saver = tf.train.Saver（max_to_keep = 1） - ＆gt; saver = tf.train.Saver（tf.trainableable_variables（），max_to_keep = 1）

然而，我仍然收到同样的错误信息。

有什么建议吗？

非常感谢，伊万

Answer 1

很抱歉可能存在混淆，第二个区块中的代码版本不正确。

这是正确的版本：

df %>% group_by(L2ID) %>% mutate(Test= ifelse(min(L2X1)==max(L2X1), TRUE,FALSE)) %>%
      mutate(L2X1_Fixed = ifelse(Test ==FALSE, min(L2X1), L2X1))

# A tibble: 6 x 8
# Groups:   L2ID [2]
   L1ID  L2ID     L1X1        L1X2  L2X1  L2X2  Test L2X1_Fixed
  <dbl> <dbl>    <dbl>       <dbl> <dbl> <dbl> <lgl>      <dbl>
1     1    11 2.355470 -1.53195614     0    13 FALSE          0
2     2    11 3.784859  0.20900278     0    13 FALSE          0
3     3    11 3.339077 -0.19772481     1    13 FALSE          0
4     4    22 2.512764  0.18222493     1     8  TRUE          1
5     5    22 1.176079  0.04175856     1     8  TRUE          1
6     6    22 3.688449 -0.42174624     1     9  TRUE          1

Tensorflow检查点不保存所有变量

1 个答案: