jupyter notebook运行问题，在c盘可以运行，在d盘就无法连接到内核

Question

为什么我使用pytorch环境的jupyter notebook在c盘运行程序时能够连接上内核，在d盘运行程序时就会显示连接不上内核

在D盘运行的时候，会一直连接内核，最后显示无法连接

在C盘运行的时候，就没问题

Answer 1

【相关推荐】

• 帮你找了个相似的问题, 你可以看下: https://ask.csdn.net/questions/379297
• 你也可以参考下这篇文章：如何在jupyter中使用安装好的pytorch
• 除此之外, 这篇博客: jupyter的一些高级设置中的 Pytorch 部分也许能够解决你的问题, 你可以仔细阅读以下内容或跳转源博客中阅读:
  

  import torch
  import math
  import torch.nn as nn
  import torch.nn.functional as F
  

  总结起来就是下面这样，注意语法别错，东西一多看着还是很头疼的，配置好就可以用了。

  {
      "snippets" : [
  	{
  		"name" : "science basic",
  		"code" : [
  			"import argparse, sys, os, errno",
  			"%pylab inline",
  			"import numpy as np",
  			"import pandas as pd",
  			"import matplotlib.pyplot as plt",
  			"plt.style.use('ggplot')",
  			"import seaborn as sns",
  			"import h5py",
  			"import os",
  			"from tqdm import tqdm",
  			"import scipy",
  			"import sklearn",
  			"from scipy.stats import pearsonr",
  			"import warnings",
  			"warnings.filterwarnings('ignore')"
  		]
          },
  	{
          	"name" : "high level plot",
          	"code" : [
                      "import matplotlib.animation as animation",
  		    "from matplotlib import rc",
  		    "from IPython.display import HTML, Image",
  		    "rc('animation', html='html5')",
  		    "import plotly",
  		    "import plotly.offline as off",
  		    "import plotly.plotly as py",
  		    "import plotly.graph_objs as go"
                  ]
          },
  	{
  		"name" : "deep learning",
  		"code" : [
  			"import keras",
  			"from keras import backend as K",
  			"from keras.callbacks import TensorBoard",
  			"from keras.callbacks import EarlyStopping",
  			"from keras.optimizers import Adam",
  			"from keras.callbacks import ModelCheckpoint",
  			"import tensorflow as tf",
  			"from keras.models import Model",
  			"from keras.layers import Input, Conv2D, MaxPooling2D, UpSampling2D,Lambda, Dot,average,add, concatenate",
  			"from keras.layers.normalization import BatchNormalization",
  			"from keras.layers.core import Dropout, Activation,Reshape",
  			"from keras.layers.merge import concatenate",
  			"from keras.callbacks import TensorBoard, EarlyStopping, ModelCheckpoint",
  			"from keras.initializers import RandomNormal",
  			"import os",
  			"os.environ['CUDA_DEVICE_ORDER'] = 'PCI_BUS_ID'",
  			"os.environ['CUDA_VISIBLE_DEVICES'] = '4'",
  			"from keras.backend.tensorflow_backend import set_session",
  			"config = tf.ConfigProto()",
  			"config.gpu_options.per_process_gpu_memory_fraction = 0.99",
  			"set_session(tf.Session(config=config))"
  			]
  		},
  	{
          	"name" : "pytorch",
          	"code" : [
                      "import torch",
  		    "import math",
  		    "import torch.nn as nn",
  		    "import torch.nn.functional as F"
                  ]
          }
      ]
  }
  

  这个配置这里坑了我很久，主要是snippets不会显示错误信息，debug得很仔细一个一个找，因此花了很久发现是keras部分的几个破双引号冲突了，还有一个注释竟然也忘删了，折腾了很久才配置好。用sublime text是可以很容易发现语法错误的，可惜最开始没在意

  本地的snippets.json在：

  /Users/james/Library/Jupyter/nbextensions/snippets/snippets.json
  

  我在james目录下也放了一份，因为最近估计需要不断更新完善，所以就放一份在：

  /Users/james/snippets.json
  

  每次修改后，需要同步到本地以及ibme、cnode、hpc1几个机器上面，这样不用我几个地方都各自改一遍，费事。

  首先找到各个机器上的json文件在哪儿

  $(jupyter --data-dir)/nbextensions/snippets/snippets.json
  

  然后写个同步的脚本syncsnip.sh

  cp /Users/james/snippets.json /Users/james/Library/Jupyter/nbextensions/snippets/snippets.json
  rsync -avzh /Users/james/snippets.json ibme:/Share/home/chenxupeng/.local/share/jupyter/nbextensions/snippets/
  rsync -avzh /Users/james/snippets.json cnode:/home/chenxupeng/.local/share/jupyter/nbextensions/snippets/snippets.json
  rsync -avzh /Users/james/snippets.json hpc1:/home/chenxupeng/.local/share/jupyter/nbextensions/snippets/
  

  然后做个同步快捷方式：

  alias snip='bash syncsnip.sh'
  

  测试几个机器的配置均通过

  配置好了还是很美的：

  

  jupyter dashboard
  看起来是可以进一步帮助展示的利器

  pip install jupyter_dashboards
  jupyter dashboards quick-setup --sys-prefix
  jupyter nbextension install --py jupyter_dashboards --sys-prefix
  jupyter nbextension enable --py jupyter_dashboards --sys-prefix
  

  不过用了一下，过分自动化的布局反而限制的比较死，，不太好看

如果你已经解决了该问题, 非常希望你能够分享一下解决方案, 写成博客, 将相关链接放在评论区, 以帮助更多的人 ^-^