Installation use
Official Document Connection: Https://www.tensorflow.org/get_started/get_started_for_beginners
In accordance with the text of the GitHub connection to download files directly GG, Hung ladder or clone do not move, helpless, had to go to that page to use the example of the py file copy came to the local, need to copy two files:
https://github.com/tensorflow/models/tree/master/samples/core/get_started/iris_data.py
https://github.com/ tensorflow/models/tree/master/samples/core/get_started/premade_estimator.py
Run directly, suggesting that there is no pandas module, after using PIP installation, the operation is correct. in the first example, the official document briefly introduces: the concept of machine learning points out that Iris problem is a classification problem this paper introduces the concept of training set and test set in machine learning with supervised learning and unsupervised learning, as well as some tips to be paid attention to during the construction. This paper introduces a function TensorFlow.Keras.utils.get_file () introduces the simple concept of the feature and the basic concept of the neural network. This paper describes the basic use of a classifier dnnclassifier, including: initialization, training, evaluation, prediction. PostScript
For people who have absolutely no access to machine learning and Python, this article and the introduction of the code is still a bit obscure, the details are no longer one by one written, the official article written in detail, it is best to read the code together, many functions need to check the use of documents.