Build a Neural Network with Pytorch - PART 1

In this lesson, I will demonstrate how to build your own neural network with Pytorch and word embedding!
This network is designed to create a story based-off several books that I've trained it on, and therefore I named it StoryTeller.
In this video, you will learn how to manually batch data, create a neural network class, and define the structure of your model.
I'll explain my trail of thought and the considerations when building such a model, and I'll also introduce you to Pytorch, which is a very powerful tool for Machine Learning and AI.
This is just an introduction, as I didn't include validation or pre-processing of our dataset, which are also very important steps that I've skipped to save some time (YES, it takes more than 35 minutes to build a proper model!)
I highly recommend downloading/copying the starter notebook from my Google Drive, and proceeding to Part 2 of this lesson once you're done structuring your model.
#pytorch #python #neuralnetwork #ai

⭐Train a Neural Network - StoryTeller PART 2 ⭐

********************************************************************************
LEARN ABOUT BASIC AI CONCEPTS - FOR MACHINE LEARNING BEGINNERS:
********************************************************************************

⭐My Perceptron Tutorial - a single node in a Neural Network ⭐

⭐My Error Function Tutorial - Binary Cross-Entropy Loss ⭐

Starter Notebook and Files:

Timestamps:
0:00 - Introduction
1:41 - Convert dataset from strings to integers
3:00 - Word to Index
4:58 - Batch Data / Train Data Structure
7:47 - Neural Network Structure
8:44 - __init__ Function
11:32 - Forward Function
14:52 - Print Model Structure
15:47 - See you in part 2!

Connect on LinkedIn:

Follow me on Facebook:

Follow me on Instagram:

Check out my Github:

[inspisiavideos]

Hey I just found your videos. They are great! I just finished my Associates degree but I learned more by watching vids like this one.

[ManojKumar-tuts]

hi Mariya, can you explain what does happen in the backend, I mean when we use functions like train etc we just see that the machine gets trained from the data but what calculations will happen inside and what does the computer do to understand the data set(we never try to learn about it). please explain this because the pytorch usage or else using tensorflow is being done by many people but many lack the knowledge of the backend which is good to know so please try to do some videos on this concept .


I also suggest you that don't do this concepts until you hit some larger number of subscribers because people are being obsessed with frameworks and forgetting to know how are those made, so these might not get better reach now but when you have a good amount of subscribers then it would give you a good value

forgot to mention your videos are great : )

[Mr-Casko]

Hey you look diferent???.... your a verry beatifull person ...God Bless ...please excuse my prior opinions i think your a valuable asset and i apreciate your resilience... thanks for helping me learn more also 🤙

[Coco]

Holy cow, this girl is gorgeous and brilliant. Thank God for discover this channel. Gretings from Mexico. 🙋‍♂️

[Shifterwizard]

Could you provide a breakdown of neural networks in general, on the coding side, similar to your OOP/classes video? One where you go over the basic parts of an individual neural network, one at a time, and common variations or pitfalls, in that OOP/classes video's visual style?

[future_philosopher]

Hello! I need some help with
data = coraline + alice + brothers_grimm
data[:10]

this is the output i'm getting:
['parent',
'indirectobject3',
'0',
'139674796863440',
'resources',
'procset',
'pdf',
'text',
'extgstate',
'indirectobject33']

not sure what's going on.. would appreciate any help!
Thank you for the amazing videos you make.

[narnittetube]

I watch all your videos and I really enjoy your level of understanding.
Can you share your study background with

[lakeguy]

I think you are using the phrase "batch_size" when you mean input_features (X subscript 0 ... X subscript 4)... you are using the previous 5 words to predict the 6th word in a sequence. (I enjoy your videos. keep up the good work!)

[akaabdullah]

Hello mam just found your channel randomly but surely one of the best channel recomended by youtube

[grahamhilton]

I have just finished an MSc in Advanced Computer Science and love your videos :)

[MisterYoupinet]

Deserves more views. thanks for sharing

[richarddiaz]

You know that’s crazy I had to make flashcards a lot of a lot of flashcards to remember projects or words thank you … but I do love to play with my circuit board yeah I know how many raspberry pies I buy computers I buy it’s a lot lol

[gaatutube]

This lecture assumes you already know what pytorch is, what embedding_dim and linear1 linear2 etc. mean. There is no explanation of these terms in this lecture. :-( I might come back some day and watch this video again once I have the preliminary mathematical understandings. I do appreciate all the hard-work you are putting into making these videos (and I have enjoyed several other videos you have made ... but this clearly out of my league for now).

[magowiz]

I wasn't able to use your notebook, the problem is with file paths, it doesn't seem to resolve them, even escaping spaces in "My Drive" with backslash.

[consciencedimension]

I think this is not needed to pass something in super() when inheriting a class. super().__init()__ should be enough

[sentrygunner]

Hey do you have a Facebook group or something? The way you teach is pretty understandable and concise! Subbed!

[apurvakumardani]

Can you convert python code into apk with buildozer for me? My lappy shows errors there. I need help

[DEFTNoah]

do you have any experience with image classification using VGG16 or other CNN models?

[shanedogg]

Finallyy someone who can get to the meats and potatoess! Best video I've seen!

[aries-eganmacdgbbit]

baby. Can you be my personal teacher. I am looking for a teacher. but i don't understand english. I have to use translation