TensorFlow 2.0 Beta has been released after Google announced TensorFlow 2.0 Alpha at the TensorFlow Developer Summit in March. TF2.0 uses Keras by default, Eager Execution, cross-platform support, and simplified APIS compared to 1.x. This update brings TF2.0 closer to PyTorch, and a series of annoying concepts will never return. If TF starts in the second half of 2019, then you will have chosen the best time to enter AI, as the Tensorflow community is thriving and looking forward to the future. Next, I will give you a detailed and complete installation tutorial for THE GPU version under TF 2.0 Beta – Window.

Directory \

1.Anaconda\

  1. Installation Anaconda
  2. Modify the path
  3. Modifying the Default Browser

2. CUDA10.0

  1. CUDA installation
  2. CuDNN installation
  3. The PATH configuration

3.TensorFlow2.0 beta-gpu version installation and testing

  1. Confirm the graphics card
  2. test

1. Anaconda

Download a.

First go to Anaconda’s website: \

www.anaconda.com/distributio…

Select Python3.7 for Windows (note: 64-bit must be selected because TF does not support python 32-bit)

Download it, open it, and install it foolishly, all the way to Next. \

B. Modify the path

The default address is drive C (this is the default address, you can skip this step if you usually install on drive C),

[jupyter_notebook_config.py] [jupyter_notebook_config.py]

Open CMD and type

jupyter notebook –generate-config

Press Enter to produce [jupyter_notebook_config.py]

Open [jupyter_notebook_config.py] with Notepad++ and find c. notebook

Establish your new work path

Uncomment the # before the c

Click Save, and now you have the path

CMD, type [jupyter notebook] and you will find that your path has been changed

C. Modify the default browser

Open the \ [jupyter_notebook_config. Py]

Find the browser path you want to use (here is my browser path)

Open [jupyter_notebook_config.py] to app. browser = “and add the following three lines below

import webbrowser

webbrowser.register(“chrome”,None,webbrowser.GenericBrowser(u”C:\ProgramFiles (x86)\Google\Chrome\Application\chrome.exe”))

c.NotebookApp.browser = ‘chrome’

It’s easy to change the browser and path used by Anaconda. Now open our Jupyter Notebook(Tensorflow2.0 notes will be written in this folder later)

2. CUDA 10.0

A. CUDA installation

Download the CUDA

Official website link:

Developer.nvidia.com/cuda-10.0-d…

After the download is complete, open the downloaded driver

Take hook GeForce Experience

If you already have Visual Studio Integration on your computer, uncheck this to avoid conflicts

Click on Driver comonents and Display Driver. CUDA itself contains a Driver version of 411.31

If the driver version on your computer is newer than the one that comes with CUDA, be sure to check this box. Otherwise, the installation will fail (in the same case, do not need to check).

After a few minutes of setup, this is the interface that the NVIDIA program has completed

Open this path and view nvcc.exe

The presence of nvcc.exe indicates a successful CUDA installation

Open this folder and check for cuti64_100.dll

CUPT1 is successful if cuti64_100. DLL is present

B. cuDNN installation

CuDNN official website link:

Developer.nvidia.com/rdp/cudnn-d…

Select cuDNN for Cuda10.0\

Unpack the cuDNN

Copy the decompressed files to CUDA folder \

C. the PATH configuration

View the CUDA environment path

My Computer — > Properties — > Advanced System Settings — > Environment Variables

Find PATH in the system variable

Check the CUDA path, which will add these two directories when you are finished installing CUDA

CUPTA and cuDNN have not been added yet, so they must be added to the path so that Tensorflow does not report errors

Add CUPTA and CUDNN paths

New — > Browse to find the path

CuDNN path, CUPTA path

CUDA for testing:

cmd

nvcc -V

The following display shows that our CUDA version is 10.0

3. TensorFlow 2.0 installation and testing

A. Confirm the graphics card

Make sure the graphics card is NVDIA before installing

The command line

PIP install tensorflow – gpu = = 2.0.0 – beta0

B. test

Tensorflow is installed successfully.

Steps:

Open the CMD – > ipython – >

import tensorflow as tf

tfabab.test.is_gpu_available()

If True is displayed, the GPU version is successfully installed

Author’s official account:

Site introduction ↓↓↓

ID: 92416895\

Currently, the planet of Knowledge in the direction of machine learning ranks no. 1.

Past wonderful review \

  • Conscience Recommendation: Introduction to machine learning and learning recommendations (2018 edition) \

  • Github Image download by Dr. Hoi Kwong (Machine learning and Deep Learning resources)

  • Printable version of Machine learning and Deep learning course notes \

  • Machine Learning Cheat Sheet – understand Machine Learning like reciting TOEFL Vocabulary

  • Introduction to Deep Learning – Python Deep Learning, annotated version of the original code in Chinese and ebook

  • The mathematical foundations of machine learning

  • Machine learning essential treasure book – “statistical learning methods” python code implementation, ebook and courseware

  • Blood vomiting recommended collection of dissertation typesetting tutorial (complete version)

  • Installation of Python (Anaconda+Jupyter Notebook +Pycharm)

  • What if Python code is ugly? Recommend a few artifacts to save you

  • Blockbuster | complete AI learning course, the most detailed resources arrangement!

Note: This site’S QQ group: 865189078 (a total of 8 groups, do not add repeatedly).

To join the wechat group of this site, please add the assistant wechat of Huang Bo, explanation: public number user group.