机器学习1

TensorBoard: 为了方便TensorFlow程序的理解、调试与优化，于是有了TensorBoard这样的可视化工具
tensorboard --logdir="./": 将日志存储到./

错误：

Bad key "text.kerning_factor" on line 4 in
/home/sunhongxuan/anaconda3/envs/tf_gpu_env/lib/python3.6/site-packages/matplotlib/mpl-data/stylelib/_classic_test_patch.mplstyle.
You probably need to get an updated matplotlibrc file from
http://github.com/matplotlib/matplotlib/blob/master/matplotlibrc.template
or from the matplotlib source distribution
1
2
3
4
5

原因：
一个名为"text.kerning_factor"的key失效引起的版本冲突，我理解的是matplotlib旧版本需要这个key，新版本不需要了；
解决方案：
根据提示找到这个key并注释掉

错误：

UserWarning: NumPy 1.14.5 or above is required for this version of SciPy (detected version 1.13.1)
  UserWarning)
1
2

原因：
Numpy版本不匹配，更新版本之后tensorboard版本有不匹配
解决方案：
全部卸载，重新下载所有内容：
CUDA:
查看推荐版本：ubuntu-drivers devices

sunhongxuan@sunhongxuan-MateBook-X-Pro:~$ ubuntu-drivers devices
== /sys/devices/pci0000:00/0000:00:1c.0/0000:01:00.0 ==
modalias : pci:v000010DEd00001D52sv00001E83sd00003E21bc03sc02i00
vendor   : NVIDIA Corporation
model    : GP108BM [GeForce MX250]
manual_install: True
driver   : nvidia-driver-450-server - distro non-free
driver   : nvidia-driver-515-server - distro non-free
driver   : nvidia-driver-418-server - distro non-hfree
driver   : nvidia-driver-470-server - distro non-free
driver   : nvidia-driver-510 - distro non-free
driver   : nvidia-driver-515 - distro non-free recommended
driver   : nvidia-driver-470 - distro non-free
driver   : nvidia-driver-510-server - distro non-free
driver   : xserver-xorg-video-nouveau - distro free builtin
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

此时终端输入sudo ubuntu-drivers autoinstall即可自动安装，或者输入sudo apt install nvidia-driver-515安装，然后重启系统即可。
nvidia--smi查看CUDA版本为:

Thu Oct 13 16:10:27 2022       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 515.65.01    Driver Version: 515.65.01    CUDA Version: 11.7     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  NVIDIA GeForce ...  Off  | 00000000:01:00.0 Off |                  N/A |
| N/A   58C    P8    N/A /  N/A |      4MiB /  2048MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|    0   N/A  N/A      1195      G   /usr/lib/xorg/Xorg                  4MiB |
+-----------------------------------------------------------------------------+
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

但是tensorflow最高支持到CUDA11.2

tensorflow-gpu版本安装失败，安装cpu版本

TensorFlow 2.6.0
python 3.6
keras 2.6.0
numpy 1.19.5
1
2
3
4

问题：

AttributeError: module 'keras.utils' has no attribute 'to_categorical'
1

解决方案：
将from keras import utils改为from keras.utils import np_utils，然后将文件中的until改为np_utils

问题：

libGL error: MESA-LOADER: failed to open iris: /usr/lib/dri/iris_dri.so: 无法打开共享对象文件: 没有那个文件或目录 (search paths /usr/lib/x86_64-linux-gnu/dri:\$${ORIGIN}/dri:/usr/lib/dri, suffix _dri)
libGL error: failed to load driver: iris
libGL error: MESA-LOADER: failed to open iris: /usr/lib/dri/iris_dri.so: 无法打开共享对象文件: 没有那个文件或目录 (search paths /usr/lib/x86_64-linux-gnu/dri:\$${ORIGIN}/dri:/usr/lib/dri, suffix _dri)
libGL error: failed to load driver: iris
libGL error: MESA-LOADER: failed to open swrast: /usr/lib/dri/swrast_dri.so: 无法打开共享对象文件: 没有那个文件或目录 (search paths /usr/lib/x86_64-linux-gnu/dri:\$${ORIGIN}/dri:/usr/lib/dri, suffix _dri)
libGL error: failed to load driver: swrast
1
2
3
4
5
6

出现这种情况很可能是核显和独显同时使用了，在ubuntu下的解决方案是：

1.确定自己安装了显卡驱动（终端输入nvidia-smi，可以查看是否有输出信息，如果有就是安装了）

2.终端输入nvidia-settings，弹出nvidia settings设置窗口，点击左侧的最后一项PRIME Profiles，选择NVIDIA（Performance Mode），然后重启。

问题：

findfont: Font family ['sans-serif'] not found. Falling back to DejaVu Sans.
1

原因：
因为是在ubuntu环境下运行的，所以没有windows中的黑体字，下载后就可以了
其实就是matplotlib输出的图片中中文乱码
解决方案：
参考博客：点我点我

警告：

2022-10-20 15:42:48.204776: E tensorflow/stream_executor/cuda/cuda_driver.cc:271] failed call to cuInit: CUDA_ERROR_SYSTEM_DRIVER_MISMATCH: system has unsupported display driver / cuda driver combination
2022-10-20 15:42:48.204907: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:169] retrieving CUDA diagnostic information for host: sunhongxuan-MateBook-X-Pro
2022-10-20 15:42:48.204933: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:176] hostname: sunhongxuan-MateBook-X-Pro
2022-10-20 15:42:48.205149: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:200] libcuda reported version is: 510.85.2
2022-10-20 15:42:48.205231: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:204] kernel reported version is: 515.76.0
2022-10-20 15:42:48.205255: E tensorflow/stream_executor/cuda/cuda_diagnostics.cc:313] kernel version 515.76.0 does not match DSO version 510.85.2 -- cannot find working devices in this configuration
2022-10-20 15:42:48.206055: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
1
2
3
4
5
6
7
8

原因：
表明 CPU 支持AVX AVX2 （可以加速CPU计算），但是你安装的 TensorFlow 版本不支持
解决方案：

1. 不管他
2. 可以卸载原来版本的tensorflow，安装cuda和cudnn和tensorflow对应的版本
   采用方案1，防止再次崩溃