Introduction To Machine Learning

Question 1

Machine learning models are computer algorithms that use data to make estimations or educated guesses and decisions.
Machine learning models differ from traditional algorithms in how they are designed.
When normal computer software needs to be improved people edit it for stop by contrast a machine-learning algorithm uses data to get better at a specific task

Question 2

Machine learning models.
The model is the core components of machine learning and ultimately what we are trying to build.
A model might estimate how old a person is from a photo predict what you might like to see on social media or decide where a robotic arm should move

Question 3

You can think of a model as a function that accept starter as an input and produces an output.
More specifically and model uses input data to estimate something else

Question 4

Parameters are discovered during training.
The human design it does not select parameter values. Instead parameter values are set to an initial guess then adjusted during an automated learning process called training.

Question 5

The goal of training is to improve a model so that it can make high-quality estimations or predictions.
Once trained you can use a model in the real world like normal software.
Models do not train themselves. They are trained using data plus two pieces of code the objective function and the optimizer.

Question 6

The objective.
The objective is what you want tomorrow to be able to do for example the objective of a scenario could be to estimate a dog’s website based on the harness size

Question 7

So that a computer can understand our objective we need to provide our goal as code snippet called and object to function also known as a cost function.
Objective function is Judge whether the models doing a good job or a bad job

Question 8

The daughter of.
Data refers to the information that we provide to the model also known as inputs.
Data also refers to information that the objective function like need. For sample if I object to function reports with the model test the boot size correctly it will need to know the correct boot size

Question 9

The optimizer.
During training the model makes their production and the objective function calculator while it sofia.
The optimizer skirt that then changes the models parameters so the model will do a better job next time stop
How and optimise it does this is complex and although we don’t normally write her own optimises we use open source frameworks where the hard work has been done for us

Question 10

It is important to keep in mind that the objective data and optimizer are simply a means to train in the model.
They are not needed once training is complete. It is also important to remember that the training for amateurs only change the parameter values inside of a model that does not change the kind of model is used

Question 11

Training vs using a model.
It is important to make a distinction between training and using a model force of new line using a model means providing inputs and receiving an estimation of production. We do this first when returning a model and when we are or our customers use it in the real world first. Using a model normally only takes less than a few seconds

Question 12

By contrast training and model is the process of improving our landlord or woolworths.
Training requires that we use the model as well as the object to function and optimizer in a special loop.
this can take minutes or days to complete.
Usually we only trainer model once. Once it’s trained we can use it as many times as we like without making further changes

Question 13

During training the objective function usually needs to know both the models out with and what’s the correct answer was these are called labels

Question 14

When we use our model we only need the columns of the data that the model accept is input.
These columns are called features.

Question 15

Taken together this means that to use a model we only ever need features well during training we usually need both features and labels

Question 16

Once a model has finished training you can save it to a file by itself.
We no longer need the original data the objective function or the model of data.
When we want to use the model we can load it from disc provided with new data and get back a prediction

Question 17

This is all of machine learning is to find patterns in data and then use these persons to make estimate.
Machine learning different from normal software development and let me use special code rather than our own intuition to improve our while the software works.

Question 18

The learning process conceptually uses for components:
Data about the topic we are interested in.
And model which makes estimates.
And objective of the model is trying to achieve.
And optimizer which is the extraco that changes the model depending on its performance.

Question 19

Data can be thought of as features and labels.
Features correspond to potential model inputs for labels respond to model outfits or desired model outputs

Question 20

Sanders and plotly are powerful tools to explore dataset in python.
Once you have trained a model we can save to this for later use