PyTorch Optimizers

[Beto-Garcia]

Description

New Concept Entry Created: content/pytorch/concepts/optimizers/optimizers.md

Issue Solved

Closes #5234

Type of Change

• Adding a new entry

Checklist

• All writings are my own.
• My entry follows the Codecademy Docs style guide.
• My changes generate no new warnings.
• I have performed a self-review of my own writing and code.
• I have checked my entry and corrected any misspellings.
• I have made corresponding changes to the documentation if needed.
• I have confirmed my changes are not being pushed from my forked main branch.
• I have confirmed that I'm pushing from a new branch named after the changes I'm making.
• I have linked any issues that are relevant to this PR in the Issues Solved section.

[cigar-galaxy82]

Suggested change

	Title: 'Optimizers' # Required; the file name should be the same as the title, but lowercase, with dashes instead of spaces, and all punctuation removed
	Description: 'A PyTorch optimizer is a tool that helps with the process of training a machine learning model. This involves adjusting the parameters of the model during training in order to minimize the error between the predicted output and the actual output.' # Required; ideally under 150 characters and starts with a noun (used in search engine results and content previews)
	Subjects: # Please only use Subjects in the subjects.md file (https://github.com/Codecademy/docs/blob/main/documentation/subjects.md). If that list feels insufficient, feel free to create a new Subject and add it to subjects.md in your PR!
	- 'How to use an optimizer'
	- 'Constructing it'
	- 'Per-parameter options'
	Tags: # Please only use Tags in the tags.md file (https://github.com/Codecademy/docs/blob/main/documentation/tags.md). If that list feels insufficient, feel free to create a new Tag and add it to tags.md in your PR!
	- 'PyTorch'
	- 'Optimizers'
	CatalogContent: # Please use course/path landing page slugs, rather than linking to individual content items. If listing multiple items, please put the most relevant one first
	- 'Concept Entry'
	- 'content/pytorch'
	Title: 'Optimizers'
	Description: 'Helps adjust the model parameters during training to minimize the error between the predicted output and the actual output.'
	Subjects:
	- 'Data Science'
	- 'Machine Learning'
	- 'AI'
	Tags:
	- 'Deep Learning'
	- 'Libraries'
	- 'TensorFlow'
	CatalogContent:
	- 'intro-to-py-torch-and-neural-networks'
	- 'paths/data-science'
	- 'paths/machine-learning'

[cigar-galaxy82]

Suggested change

	**torch.optim** is a package implementing various optimization algorithms.
	Most commonly used methods are already supported, and the interface is general enough, so that more sophisticated ones can also be easily integrated in the future.
	Optimizers help adjust the model parameters during training to minimize the error between the predicted output and the actual output. They use the gradients calculated through backpropagation to update the model in a direction that reduces this error, improving the model's performance over time.

[cigar-galaxy82]

Suggested change

	## How to use an optimizer
	To use ```torch.optim``` you have to construct an optimizer object that will hold the current state and will update the parameters based on the computed gradients.
	## Constructing it
	To construct an Optimizer you have to give it an iterable containing the parameters (all should be Variable s) to optimize. Then, you can specify optimizer-specific options such as the learning rate, weight decay, etc.
	Example:
	```codebyte/js
	optimizer = optim.SGD(model.parameters(), lr=0.01, momentum=0.9)
	optimizer = optim.Adam([var1, var2], lr=0.0001)
	```
	## How to use an optimizer
	To use an optimizer including the `torch.optim` package is a must and constructing an optimizer object that will hold the current state and will update the parameters based on the computed gradients.
	## Syntax
	```pseudo
	# Importing the package
	import torch.optim as optim
	# Instantiate the model that will be trained
	model = YourModel()
	# Instantiate the optimizer of choice
	optimizer = optim.optimizer_name()
	# Training the model
	for epoch in range(num_epochs):
	# Steps for training the model using an optimizer

[Sriparno08]

Looks good, @Beto-Garcia!

[github-actions]

👋 @Beto-Garcia
You have contributed to Codecademy Docs, and we would like to know more about you and your experience.
Please take a minute to fill out this four question survey to help us better understand Docs contributions and how we can improve the experience for you and our learners.
Thank you for your help!

🎉 Your contribution(s) can be seen here:

https://www.codecademy.com/resources/docs/pytorch/optimizers

Please note it may take a little while for changes to become visible.
If you're appearing as anonymous and want to be credited, see here.