In our previous section we have seen how a machine can learn from users. We call these examples training set in machine learning.
We indicate each training example as a training instance.
As we have seen earlier, as users mark emails as spams, and move them to spam folders, the system uses them as examples.
For example, our spam filter becomes a machine learning program that learns to flag spam thereafter.
However, it is a continuous process. And the software learns to separate regular and spam emails quite correctly.
Broadly speaking, we can train a model in two ways.
The first one is supervised learning.
What is Supervised Learning?
Before defining supervised learning, let us know one key concept. A training set has an input and it gives us an output. Right?
Users’ experience works here as inputs, of course.
As a result, the input and output build a connection. We can call each of them as variables.
If you are a complete beginner your journey to learn TensorFlow might start from here.
For the TensorFlow beginners we have a dedicated category – TensorFlow for Beginners.
But besides that, you may need to learn several other machine learning and data science libraries.
As a result, you may check these categories as well – NumPy, Pandas, Matplotlib.
However, without learning Python, you cannot learn the usages of these libraries. Why? Because they all use Python as the Programming language.
Therefore please learn Python at the very beginning and start learning TensorFlow.
And, finally please check our Mathematics, Discrete Mathematics and Data Structures categories specially. We have tried to discuss from basic to intermediate level so that you can pick up the core ideas of TensorFlow.
To distinguish properly we can attach labels to put them in order.
As an outcome, reading those labels machines can learn faster.
This is supervised learning.
Certainly, there always exists a common pattern, based on which the machine predicts the output.
What is Unsupervised Learning
Just the opposite. The unsupervised learning does not require the labels for training sets.
As an outcome, models don’t need any label for reading and finding the common pattern, and making predictions.
However, with reference to unsupervised learning, models heavily depend on past predictions.
By the way, we call the training set a training dataset, or taring data.
In the future, we will use the training set in machine learning to understand these concepts more deeply.
Leave a Reply