window 安装 PyTorch

太过爱你忘了你带给我的痛 2021-12-19 14:59 411阅读 0赞

### ![20191009191333910.png][]**[日萌社][Link 1]** ###

**[人工智能AI：Keras PyTorch MXNet TensorFlow PaddlePaddle 深度学习实战（不定时更新）][AI_Keras PyTorch MXNet TensorFlow PaddlePaddle]**

--------------------

[Ubuntu 安装 CUDA、cuDNN、anaconda、pycharm、tensorflow、pytorch、JDK1.8][Ubuntu _ CUDA_cuDNN_anaconda_pycharm_tensorflow_pytorch_JDK1.8]

[window下 人工智能 Keras、TensorFlow、PyTorch、CUDA、cuDNN 的环境安装 总文章][window_ _ Keras_TensorFlow_PyTorch_CUDA_cuDNN _]

[window 安装 PyTorch][window _ PyTorch]

[人工智能AI：Keras PyTorch MXNet 深度学习实战（不定时更新）][AI_Keras PyTorch MXNet TensorFlow PaddlePaddle]

--------------------

[Ubuntu 安装 CUDA、cuDNN、anaconda、pycharm、tensorflow、pytorch、JDK1.8][Ubuntu _ CUDA_cuDNN_anaconda_pycharm_tensorflow_pytorch_JDK1.8 1]

[PyTorch 安装 GPU版本(CUDA/cuDNN)][PyTorch _ GPU_CUDA_cuDNN]

[PyTorch 安装 GPU版本(CUDA/cuDNN)][PyTorch _ GPU_CUDA_cuDNN 1]

[window 安装 PyTorch][window _ PyTorch 1]

--------------------

pytorch gpu 安装
    	1.官网下载安装：https://pytorch.org/get-started/locally/
    	  安装命令：conda install pytorch torchvision cudatoolkit=10.1 -c pytorch
    	2.国内源直接下载链接
    		网站链接：
                        window：https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/win-64/
                        linux：https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/linux-64/
    		选择下载：
                        window：pytorch-1.5.0-py3.7_cpu_0.tar.bz2
                        linux：pytorch-1.5.0-py3.7_cuda10.1.243_cudnn7.6.3_0.tar.bz2
    		下载链接：
                        window：https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/win-64/pytorch-1.5.0-py3.7_cpu_0.tar.bz2
                        linux：https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/linux-64/pytorch-1.5.0-py3.7_cuda10.1.243_cudnn7.6.3_0.tar.bz2
    		安装：
                        window：conda install pytorch-1.5.0-py3.7_cpu_0.tar.bz2
                        linux：conda install pytorch-1.5.0-py3.7_cuda10.1.243_cudnn7.6.3_0.tar.bz2　　　
    		pip list
    			torch  1.5.0
    	3.测试GPU版本的代码
    		import torch
    		#测试 CUDA
    		print("Support CUDA ?: ", torch.cuda.is_available()) #返回True
    		x = torch.tensor([10.0])
    		x = x.cuda()
    		print(x)
    		y = torch.randn(2, 3)
    		y = y.cuda()
    		print(y)
    		z = x + y
    		print(z)
    		# 测试 CUDNN
    		from torch.backends import cudnn
    		print("Support cudnn ?: ",cudnn.is_acceptable(x)) #返回True

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3ppbWlhbzU1MjE0NzU3Mg_size_16_color_FFFFFF_t_70][]

--------------------

1.安装链接：https://pytorch.org/get-started/locally/
      window下装的软件版本：CUDA 9.0、cuDNN 7.1.4、Python 3.6.5
      conda 在线安装命令：conda install pytorch torchvision cudatoolkit=9.0 -c pytorch
      pip3 在线安装命令：pip3 install https://download.pytorch.org/whl/cu90/torch-1.1.0-cp36-cp36m-win_amd64.whl
    	            pip3 install https://download.pytorch.org/whl/cu90/torchvision-0.3.0-cp36-cp36m-win_amd64.whl

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3ppbWlhbzU1MjE0NzU3Mg_size_16_color_FFFFFF_t_70 1][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3ppbWlhbzU1MjE0NzU3Mg_size_16_color_FFFFFF_t_70 2][]

2.测试一下是否安装好
    		1.输入 import torch
    		  输出 Traceback (most recent call last):
    		  	File "<ipython-input-79-eb42ca6e4af3>", line 1, in <module>
    				import torch
    		  	File "G:\Anaconda3\lib\site-packages\torch\__init__.py", line 79, in <module>
    				from torch._C import *
    			ModuleNotFoundError: No module named 'numpy.core._multiarray_umath'
    		  解决：查看numpy的当前版本 conda list numpy 或 pip show numpy
    		        更新numpy的版本 pip install --upgrade numpy 或 pip install -i https://pypi.tuna.tsinghua.edu.cn/simple --upgrade numpy
    
    		2.更新numpy的版本报错：Consider using the `--user` option or check the permissions.
    		  解决：pip install --user --upgrade  numpy

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3ppbWlhbzU1MjE0NzU3Mg_size_16_color_FFFFFF_t_70 3][]

3.查看pytorch版本
    	import torch
    	print(torch.__version__)
    	1.1.0
    
    4.看看能不能用GPU加速：
    	import torch
    	torch.__version__
    	Out[3]: '1.1.0'
    	torch.cuda.is_available()
    	Out[4]: True

[20191009191333910.png]: /images/20211219/43483886c6374693a472903a420ac418.png
[Link 1]: http://www.rimengshe.com/
[AI_Keras PyTorch MXNet TensorFlow PaddlePaddle]: https://blog.csdn.net/zimiao552147572/article/details/88867161
[Ubuntu _ CUDA_cuDNN_anaconda_pycharm_tensorflow_pytorch_JDK1.8]: https://blog.csdn.net/zimiao552147572/article/details/105821003
[window_ _ Keras_TensorFlow_PyTorch_CUDA_cuDNN _]: https://blog.csdn.net/zimiao552147572/article/details/88854126
[window _ PyTorch]: https://blog.csdn.net/zimiao552147572/article/details/94333706
[Ubuntu _ CUDA_cuDNN_anaconda_pycharm_tensorflow_pytorch_JDK1.8 1]: https://mp.csdn.net/console/editor/html/105821003
[PyTorch _ GPU_CUDA_cuDNN]: https://mp.csdn.net/console/editor/html/105416761
[PyTorch _ GPU_CUDA_cuDNN 1]: https://mp.csdn.net/console/editor/html/104766646
[window _ PyTorch 1]: https://mp.csdn.net/console/editor/html/94333706
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3ppbWlhbzU1MjE0NzU3Mg_size_16_color_FFFFFF_t_70]: /images/20211219/f15af6a481a4400086b87c5af3f78cf4.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3ppbWlhbzU1MjE0NzU3Mg_size_16_color_FFFFFF_t_70 1]: /images/20211219/2271b514e08647adaa5dd3257516cf9d.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3ppbWlhbzU1MjE0NzU3Mg_size_16_color_FFFFFF_t_70 2]: /images/20211219/6b3bc143ea4b4cefa06f6b453f2525e5.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3ppbWlhbzU1MjE0NzU3Mg_size_16_color_FFFFFF_t_70 3]: /images/20211219/e69f1eb688204513b8a359ea0ff6219a.png