A super machine learning 01

Machine Learning 001: Anaconda installation

       Machine learning is my senior phase in contact with fresh material, I want to have it, study it. I learn TensorFlow framework, it is our machine learning is an important tool for online courses, books or interviews companies recommend I use Linux systems to run TensorFlow environment. All have Well, I'm still exploratory stage so I do not like to jump to Linux, the next whole chapter should all use Windows systems. ,

  • Download Anaconda

           Search official website, click on the download https://www.anaconda.com/

 

After selecting the download version of the software corresponding to the selected number of digits corresponding computer system (Apple / win / Linux) of (32/64)

The next step is to download and run the Anaconda installation package (previously in D drive new folder as Anaconda)

Installation directory to put our new folder (make sure the folder is empty, otherwise there will be a warning cause can not be installed in the specified folder)

 

I created a folder placed in Anaconda Anaconda installation package, so I need to build a new one.

Both see their own situation choice Da Buda hook hook, not a big problem. So the next step may be.

 

 

Both hooks I have canceled, as advertised, as if you really are interested can try.

 

          Here, Anaconda has been our initial build up, Anaconda is now only base environment, we need a TensorFlow environment.

          In Anaconda Prompt

               conda create --name tensorflow python=3.7 anaconda 

           After this command, you will have TensorFlow environment

 

 At least in Anaconda in your program, tool kits, are two unrelated program in base environment and TensorFlow environment, tool kits.

Open Spyder

 

 Spyder is a very handy program similar to Matlab user interface.

In the workspace enter: import tensorflow as tf there will be a warning: do not tensorflow this module,

Then you should go back to Anaconda Prompt activation tensorflow environment

 

Activate command: activate tensorflow

Exit environment: deactivate 

 

Tensorflow package mounted in tensorflow environment. Run command: conda install tensorflow

After installation, import tensorflow as tf no error (meaning of this code is coated with introducing tensorflow do tensorflow alias tf)

 

 Benpian Up to this, the following specific instructions not run if the lack of a package, Google will be available, thanks

 

Guess you like

Origin www.cnblogs.com/Micla/p/11032744.html