非法指令：运行任何 TensorFlow 程序时出现 4 错误

Question

我正在尝试训练一个 Tensorflow 卷积神经网络，但无论我运行程序的环境如何，我总是会遇到一个神秘的错误。

在 Jupyter Notebook 中，内核直接死掉了。

在终端中，我得到“非法指令：4”，没有回溯。

在 Pycharm 中，我得到：“进程以退出代码 132 完成（被信号 4 中断：SIGILL）”。

我查看了整个 Internet，但没有找到在这种情况下引发此特定错误的任何实例。如果有人能帮助阐明这个错误，我将不胜感激。

我正在使用带有 python 3.6.2 的 Mac OS X High Sierra

我的代码可以在下面找到，正如我之前所说，没有回溯。

import tensorflow as tf
import numpy as np
import pandas as pd

# OS to load files and save checkpoints
import os

image_height = 60
image_width = 1

image1_height = 15
image2_width = 1
model_name = "tensorflowCNN"

train_data = np.asarray(pd.read_csv("/home/student/Desktop/TrainingInput.csv", usecols=[1]))
lis = train_data.tolist()
lis = lis[0:60]
lis = [x[0].strip('[]\n,') for x in lis]
nlis = []
for i in lis:
    nlis.append(i.split())
for i in range(len(nlis)):
    nlis[i] = [float(x) for x in nlis[i] if x != "...,"]
nlis = [np.mean(x) for x in nlis]

train_data = np.asarray(nlis)
train_labels = np.asarray(pd.read_csv("/home/student/Desktop/TrainingInput.csv", usecols=[2]))
mylist = train_labels.tolist()
mylist = mylist[0:60]
mylist = [x[0] for x in mylist]
for i in range(len(mylist)):
    if mylist[i] == "GravelTraining":
        mylist[i] = 1.0
    elif mylist[i] == "WaterTraining":
        mylist[i] = 2.0
    else:
        mylist[i] = 3.0
print(mylist)
train_labels = np.asarray(mylist)

eval_data = np.asarray(pd.read_csv("/home/student/Desktop/TestingInput.csv", usecols=[1]))
List = eval_data.tolist()
List = List[0:15]
eval_data = np.asarray(List)
eval_labels = np.asarray(pd.read_csv("/home/student/Desktop/TestingInput.csv", usecols=[2]))
myList = eval_labels.tolist()
myList = myList[0:15]
eval_labels = np.asarray(myList)

category_names = list(map(str, range(3)))

# TODO: Process mnist data
train_data = np.reshape(train_data, (-1, image_height, image_width, 1))
train_labels = np.reshape(train_labels, (-1, image_height, image_width, 1))
eval_labels = np.reshape(eval_labels, (-1, image1_height, image2_width, 1))
eval_data = np.reshape(eval_data, (-1, image1_height, image2_width, 1))


# TODO: The neural network
class ConvNet:

    def __init__(self, image_height, Image_width, num_classes, chan):
        self.input_layer = tf.placeholder(dtype=tf.float32, shape=[1, image_height, Image_width, chan], name="inputs")

        conv_layer_1 = tf.layers.conv2d(self.input_layer, filters=32, kernel_size=[5, 5], padding="same",
                                        activation=tf.nn.relu)
        pooling_layer_1 = tf.layers.max_pooling2d(conv_layer_1, pool_size=[2, 1], strides=1)

        conv_layer_2 = tf.layers.conv2d(pooling_layer_1, filters=64, kernel_size=[5, 5], padding="same",
                                        activation=tf.nn.relu)
        pooling_layer_2 = tf.layers.max_pooling2d(conv_layer_2, pool_size=[2,1], strides=2)

        conv_layer_3 = tf.layers.conv2d(pooling_layer_2, filters=128, kernel_size=[5,5], padding="same",
                                        activation=tf.nn.relu)
        pooling_layer_3 = tf.layers.max_pooling2d(conv_layer_3, pool_size=[2,1], strides=2)

        flattened_pooling = tf.layers.flatten(pooling_layer_1)
        dense_layer = tf.layers.dense(flattened_pooling, 60, activation=tf.nn.relu)

        dropout = tf.layers.dropout(dense_layer, rate=0.4, training=True)

        output_dense_layer = tf.layers.dense(dropout, num_classes)

        self.choice = tf.argmax(output_dense_layer, axis=1)
        self.probabilities = tf.nn.softmax(output_dense_layer)

        self.labels = tf.placeholder(dtype=tf.float32, name="labels")
        self.accuracy, self.accuracy_op = tf.metrics.accuracy(self.labels, self.choice)

        one_hot_labels = tf.one_hot(indices=tf.cast(self.labels, dtype=tf.int32), depth=num_classes)
        self.loss = tf.losses.softmax_cross_entropy(onehot_labels=one_hot_labels, logits=output_dense_layer)

        optimizer = tf.train.GradientDescentOptimizer(learning_rate=1e-2)
        self.train_operation = optimizer.minimize(loss=self.loss, global_step=tf.train.get_global_step())


# Training process:variables

training_steps = 20000
batch_size = 60

path = "./" + model_name + "-cnn/"

load_checkpoint = False

tf.reset_default_graph()
dataset = tf.data.Dataset.from_tensor_slices((train_data, train_labels))
dataset = dataset.shuffle(buffer_size=train_labels.shape[0])
dataset = dataset.batch(batch_size)
dataset = dataset.repeat()

dataset_iterator = dataset.make_initializable_iterator()
next_element = dataset_iterator.get_next()

cnn = ConvNet(image_height, image_width, 1, 1)
print("milestone1")
saver = tf.train.Saver(max_to_keep=2)
print('milestone2')
if not os.path.exists(path):
    os.makedirs(path)
print('milestone3')
with tf.Session() as sess:
    #     if load_checkpoint:
    #         print(path)
    #         checkpoint = tf.train.get_checkpoint_state(path)
    #         print(checkpoint)
    #         saver.restore(sess, checkpoint.model_checkpoint_path)
    #     else:
    sess.run(tf.global_variables_initializer())
    print('milestone4')
    sess.run(tf.local_variables_initializer())
    sess.run(dataset_iterator.initializer)
    for step in range(training_steps):
        current_batch = sess.run(next_element)
        batch_inputs = current_batch[0]
        batch_labels = current_batch[1]
        sess.run((cnn.train_operation, cnn.accuracy_op),
                 feed_dict={cnn.input_layer: batch_inputs, cnn.labels: batch_labels})

        if step % 1 == 0 and step > 0:
            current_acc = sess.run(cnn.accuracy)
            print("Accuracy at step " + str(step) + ":" + str(current_acc))
            saver.save(sess, path + model_name, step)

    print("Saving final checkpoint for training session.")
    saver.save(sess, path + model_name, step)

提前致谢。

Answer 1

好的 如果您的 2.66 GHz 版本在我看来是 2010 年发布的 Arrendale 架构，在这种情况下，它绝对没有机会工作，因为该东西没有最新的张量流二进制文件所需的 AVX 指令。

除非你的 CPU 是 Sandy Bridge 或更新的（所以 AVX 指令）

您的选择是：

1) 获得更新的 CPU
2) 安装旧版本的张量流
3) 从源代码编译张量流

降级版本见。
Illegal instruction(core dumped) tensorflow
Illegal instruction when import tensorflow in Python

编辑
似乎虽然所有 Core（i3、i5、i7）和 Xeon CPU Sandy Bridge 和更新的 CPU 都支持 AVX，但截至 2018 年，Celron 和 Pentium CPU 都不是这样。如果购买硬件，请检查机器包含的内容。