Скачать или смотреть ML Databases and How to Access them with Pytorch - MNIST Tutorial

In this tutorial, we will talk about popular Machine Learning databases and how we can easily access them with Pytorch. 🔥🔥🔥
In particular, we will focus on MNIST, which is a handwritten digits database with 70,000 different images. We will load it with a very simple Pytorch command and we will have a closer look at its content, as well as its feature + label structure.
We will also discuss data transforms, why we need them and how do we decide which transforms work better for what kind of data.

Other computer vision databases we will briefly discuss are CIFAR-10, FashionMNIST and HMDB51. You can find the full list of available databases and transforms at the very bottom of the description ⬇⬇⬇

Have you seen the previous ML tutorial I refer to in this video?
   • Neural Network Simply Explained - Deep Lea...  

New to Google Colab?
In my following video, I show you how to set it up for the first time:
   • Make Graphs with Pandas in Less Than 10 Mi...  

**************************************
⏰Time Stamps ⏰
00:00 - intro
00:23 - load MNIST with Pytorch
02:01 - MNIST features and targets
03:20 - Pytorch Databases
04:14 - data transforms
07:55 - in the next tutorial

**************************************
⭐ IMPORTS ⭐
(starter code for the entire project- not just this video)
**************************************
import torch
from torch import nn, optim
from torch.utils.data import DataLoader
from torchvision import datasets, transforms, models
from torchvision.transforms import ToTensor, Lambda, Compose
import matplotlib.pyplot as plt

**************************************
⭐ INSTALL DEPENDENCIES ⭐
⭐ ANACONDA - RECOMMENDED! ⭐

conda install -c pytorch pytorch
conda install -c pytorch torchvision

**************************************
⭐ INSTALL DEPENDENCIES ⭐
⭐ PYPI - ONLY IF THERE'S NO OTHER CHOICE! ⭐

pip install torch
pip install torchvision

**************************************
⭐ IMPORTANT LINKS ⭐

🤩 All Available Torchvision Datasets:
https://pytorch.org/vision/stable/dat...

🤩 All Available Pytorch Transforms:
https://pytorch.org/vision/stable/tra...

🤩 Complete Colab Notebook for loading MNIST, FashionMNIST and CIFAR-10:
https://colab.research.google.com/dri...

🤩 Complete Notebook for Next Lesson (Neural Network Training on MNIST Data):
https://colab.research.google.com/dri...

Please keep in mind that the next lessons notebook doesn't include testing and validation!!
**************************************

🔊 My apologies for the sound quality 🔊
I'm trying different recording solutions and this one is definitely NOT THE ONE! 😅

😍 Beautiful Icons used in this video and thumbnail are by: flaticon.com 😍