Trending March 2024 # Build Your First Image Classification Model In Just 10 Minutes! # Suggested April 2024 # Top 6 Popular

You are reading the article Build Your First Image Classification Model In Just 10 Minutes! updated in March 2024 on the website Moimoishop.com. We hope that the information we have shared is helpful to you. If you find the content interesting and meaningful, please share it with your friends and continue to follow and support us for the latest updates. Suggested April 2024 Build Your First Image Classification Model In Just 10 Minutes!

14 minutes

⭐⭐⭐⭐⭐

Rating: 5 out of 5.

Introduction

“Build a deep learning model in a few minutes? It’ll take hours to train! I don’t even have a good enough machine.” I’ve heard this countless times from aspiring data scientists who shy away from building deep learning models on their own machines.

You don’t need to be working for Google or other big tech firms to work on deep learning datasets! It is entirely possible to build your own neural network from the ground up in a matter of minutes without needing to lease out Google’s servers. Fast.ai’s students designed a model on the Imagenet dataset in 18 minutes – and I will showcase something similar in this article.

Deep learning is a vast field so we’ll narrow our focus a bit and take up the challenge of solving an Image Classification project. Additionally, we’ll be using a very simple deep learning architecture to achieve a pretty impressive accuracy score.

You can consider the Python code we’ll see in this article as a benchmark for building Image Classification models. Once you get a good grasp on the concept, go ahead and play around with the code, participate in competitions and climb up the leaderboard!

If you’re new to deep learning and are fascinated by the field of computer vision (who isn’t?!), do check out the ‘Computer Vision using Deep Learning‘ course. It’s a comprehensive introduction to this wonderful field and will set you up for what is inevitably going to a huge job market in the near future.

Project to apply Image Classification

Problem Statement

More than 25% of the entire revenue in E-Commerce is attributed to apparel & accessories. A major problem they face is categorizing these apparels from just the images especially when the categories provided by the brands are inconsistent. This poses an interesting computer vision problem that has caught the eyes of several deep learning researchers.

Fashion MNIST is a drop-in replacement for the very well known, machine learning hello world – MNIST dataset which can be checked out at ‘Identify the digits’ practice problem. Instead of digits, the images show a type of apparel e.g. T-shirt, trousers, bag, etc. The dataset used in this problem was created by Zalando Research.

Practice Now

What Is Image Classification?

Consider the below image:

You will have instantly recognized it – it’s a (swanky) car. Take a step back and analyze how you came to this conclusion – you were shown an image and you classified the class it belonged to (a car, in this instance). And that, in a nutshell, is what image classification is all about.

There are potentially n number of categories in which a given image can be classified. Manually checking and classifying images is a very tedious process. The task becomes near impossible when we’re faced with a massive number of images, say 10,000 or even 100,000. How useful would it be if we could automate this entire process and quickly label images per their corresponding class?

Now that we have a handle on our subject matter, let’s dive into how an image classification model is built, what are the prerequisites for it, and how it can be implemented in Python.

Setting Up the Structure of Our Image Data

Our data needs to be in a particular format in order to solve an image classification problem. We will see this in action in a couple of sections but just keep these pointers in mind till we get there.

You should have 2 folders, one for the train set and the other for the test set. In the training set, you will have a .csv file and an image folder:

The .csv file contains the names of all the training images and their corresponding true labels

The image folder has all the training images.

The .csv file in our test set is different from the one present in the training set. This test set .csv file contains the names of all the test images, but they do not have any corresponding labels. Can you guess why? Our model will be trained on the images present in the training set and the label predictions will happen on the testing set images

If your data is not in the format described above, you will need to convert it accordingly (otherwise the predictions will be awry and fairly useless).

Breaking Down the Process of Model Building

Before we deep dive into the Python code, let’s take a moment to understand how an image classification model is typically designed. We can divide this process broadly into 4 stages. Each stage requires a certain amount of time to execute:

Loading and pre-processing Data – 30% time

Defining Model architecture – 10% time

Training the model – 50% time

Estimation of performance – 10% time

Let me explain each of the above steps in a bit more detail. This section is crucial because not every model is built in the first go. You will need to go back after each iteration, fine-tune your steps, and run it again. Having a solid understanding of the underlying concepts will go a long way in accelerating the entire process.

Stage 1: Loading and pre-processing the data

Data is gold as far as deep learning models are concerned. Your image classification model has a far better chance of performing well if you have a good amount of images in the training set. Also, the shape of the data varies according to the architecture/framework that we use.

Hence, the critical data pre-processing step (the eternally important step in any project). I highly recommend going through the ‘Basics of Image Processing in Python’ to understand more about how pre-processing works with image data.

But we are not quite there yet. In order to see how our model performs on unseen data (and before exposing it to the test set), we need to create a validation set. This is done by partitioning the training set data.

In short, we train the model on the training data and validate it on the validation data. Once we are satisfied with the model’s performance on the validation set, we can use it for making predictions on the test data.

Time required for this step: We require around 2-3 minutes for this task.

Stage 2: Defining the model’s architecture

This is another crucial step in our deep learning model building process. We have to define how our model will look and that requires answering questions like:

How many convolutional layers do we want?

What should be the activation function for each layer?

How many hidden units should each layer have?

And many more. These are essentially the hyperparameters of the model which play a MASSIVE part in deciding how good the predictions will be.

How do we decide these values? Excellent question! A good idea is to pick these values based on existing research/studies. Another idea is to keep experimenting with the values until you find the best match but this can be quite a time consuming process.

Time required for this step: It should take around 1 minute to define the architecture of the model.

Stage 3: Training the model

For training the model, we require:

Training images and their corresponding true labels

Validation images and their corresponding true labels (we use these labels only to validate the model and not during the training phase)

We also define the number of epochs in this step. For starters, we will run the model for 10 epochs (you can change the number of epochs later).

Time required for this step: Since training requires the model to learn structures, we need around 5 minutes to go through this step.

And now time to make predictions!

Stage 4: Estimating the model’s performance

Finally, we load the test data (images) and go through the pre-processing step here as well. We then predict the classes for these images using the trained model.

Time required for this step: ~ 1 minute.

Setting Up the Problem Statement and Understanding the Data

We will be picking up a really cool challenge to understand image classification. We have to build a model that can classify a given set of images according to the apparel (shirt, trousers, shoes, socks, etc.). It’s actually a problem faced by many e-commerce retailers which makes it an even more interesting computer vision problem.

This challenge is called ‘Identify the Apparels’ and is one of the practice problems we have on our DataHack platform. You will have to register and download the dataset from the above link.

We have a total of 70,000 images (28 x 28 dimension), out of which 60,000 are from the training set and 10,000 from the test one. The training images are pre-labelled according to the apparel type with 10 total classes. The test images are, of course, not labelled. The challenge is to identify the type of apparel present in all the test images.

We will build our model on Google Colab since it provides a free GPU to train our models.

Steps to Build Our Model

Time to fire up your Python skills and get your hands dirty. We are finally at the implementation part of our learning!

Setting up Google Colab

Importing Libraries

Loading and Preprocessing Data – (3 mins)

Creating a validation set

Defining the model structure – (1 min)

Training the model – (5 min)

Making predictions – (1 min)

Let’s look at each step in detail.

Step 1: Setting up Google Colab

Since we’re importing our data from a Google Drive link, we’ll need to add a few lines of code in our Google Colab notebook. Create a new Python 3 notebook and write the following code blocks:

!pip install PyDrive

This will install PyDrive. Now we will import a few required libraries:

import os from chúng tôi import GoogleAuth from pydrive.drive import GoogleDrive from google.colab import auth from oauth2client.client import GoogleCredentials

Next, we will create a drive variable to access Google Drive:

auth.authenticate_user() gauth = GoogleAuth() gauth.credentials = GoogleCredentials.get_application_default() drive = GoogleDrive(gauth)

To download the dataset, we will use the ID of the file uploaded on Google Drive:

download = drive.CreateFile({'id': '1BZOv422XJvxFUnGh-0xVeSvgFgqVY45q'})

Replace the ‘id’ in the above code with the ID of your file. Now we will download this file and unzip it:

download.GetContentFile('train_LbELtWX.zip') !unzip train_LbELtWX.zip

You have to run these code blocks every time you start your notebook.

Step 2 : Import the libraries we’ll need during our model building phase.

import keras from keras.models import Sequential from keras.layers import Dense, Dropout, Flatten from keras.layers import Conv2D, MaxPooling2D from keras.utils import to_categorical from keras.preprocessing import image import numpy as np import pandas as pd import matplotlib.pyplot as plt from sklearn.model_selection import train_test_split from keras.utils import to_categorical from tqdm import tqdm

Step 3: Recall the pre-processing steps we discussed earlier. We’ll be using them here after loading the data.

train = pd.read_csv('train.csv')

Next, we will read all the training images, store them in a list, and finally convert that list into a numpy array.

# We have grayscale images, so while loading the images we will keep grayscale=True, if you have RGB images, you should set grayscale as False train_image = [] for i in tqdm(range(train.shape[0])):     img = image.load_img('train/'+train['id'][i].astype('str')+'.png', target_size=(28,28,1), grayscale=True)     img = image.img_to_array(img)     img = img/255     train_image.append(img) X = np.array(train_image)

As it is a multi-class classification problem (10 classes), we will one-hot encode the target variable.

y=train['label'].values y = to_categorical(y)

Step 4: Creating a validation set from the training data.

X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42, test_size=0.2)

Step 5: Define the model structure.

We will create a simple architecture with 2 convolutional layers, one dense hidden layer and an output layer.

model = Sequential() model.add(Conv2D(32, kernel_size=(3, 3),activation='relu',input_shape=(28,28,1))) model.add(Conv2D(64, (3, 3), activation='relu')) model.add(MaxPooling2D(pool_size=(2, 2))) model.add(Dropout(0.25)) model.add(Flatten()) model.add(Dense(128, activation='relu')) model.add(Dropout(0.5)) model.add(Dense(10, activation='softmax'))

Next, we will compile the model we’ve created.

Step 6: Training the model.

In this step, we will train the model on the training set images and validate it using, you guessed it, the validation set.

model.fit(X_train, y_train, epochs=10, validation_data=(X_test, y_test))

Step 7: Making predictions!

We’ll initially follow the steps we performed when dealing with the training data. Load the test images and predict their classes using the model.predict_classes() function.

download = drive.CreateFile({'id': '1KuyWGFEpj7Fr2DgBsW8qsWvjqEzfoJBY'}) download.GetContentFile('test_ScVgIM0.zip') !unzip test_ScVgIM0.zip

Let’s import the test file:

test = pd.read_csv('test.csv')

Now, we will read and store all the test images:

test_image = [] for i in tqdm(range(test.shape[0])):     img = image.load_img('test/'+test['id'][i].astype('str')+'.png', target_size=(28,28,1), grayscale=True)     img = image.img_to_array(img)     img = img/255     test_image.append(img) test = np.array(test_image) # making predictions prediction = model.predict_classes(test)

We will also create a submission file to upload on the DataHack platform page (to see how our results fare on the leaderboard).

download = drive.CreateFile({'id': '1z4QXy7WravpSj-S4Cs9Fk8ZNaX-qh5HF'}) download.GetContentFile('sample_submission_I5njJSF.csv') # creating submission file sample = pd.read_csv('sample_submission_I5njJSF.csv') sample['label'] = prediction sample.to_csv('sample_cnn.csv', header=True, index=False)

Download this sample_cnn.csv file and upload it on the contest page to generate your results and check your ranking on the leaderboard. This will give you a benchmark solution to get you started with any Image Classification problem!

You can try hyperparameter tuning and regularization techniques to improve your model’s performance further. I ecnourage you to check out this article to understand this fine-tuning step in much more detail – ‘A Comprehensive Tutorial to learn Convolutional Neural Networks from Scratch’.

New Practice Problem

Let’s test our learning on a different dataset. We’ll be cracking the ‘Identify the Digits’ practice problem in this section. Go ahead and download the dataset. Before you proceed further, try to solve this on your own. You already have the tools to solve it – you just need to apply them! Come back here to check your results or if you get stuck at some point.

In this challenge, we need to identify the digit in a given image. We have a total of 70,000 images – 49,000 labelled ones in the training set and the remaining 21,000 in the test set (the test images are unlabelled). We need to identify/predict the class of these unlabelled images.

Ready to begin? Awesome! Create a new Python 3 notebook and run the following code:

# Setting up Colab !pip install PyDrive import os from chúng tôi import GoogleAuth from pydrive.drive import GoogleDrive from google.colab import auth from oauth2client.client import GoogleCredentials auth.authenticate_user() gauth = GoogleAuth() gauth.credentials = GoogleCredentials.get_application_default() drive = GoogleDrive(gauth) # Replace the id and filename in the below codes download = drive.CreateFile({'id': '1ZCzHDAfwgLdQke_GNnHp_4OheRRtNPs-'}) download.GetContentFile('Train_UQcUa52.zip') !unzip Train_UQcUa52.zip # Importing libraries import keras from keras.models import Sequential from keras.layers import Dense, Dropout, Flatten from keras.layers import Conv2D, MaxPooling2D from keras.utils import to_categorical from keras.preprocessing import image import numpy as np import pandas as pd import matplotlib.pyplot as plt from sklearn.model_selection import train_test_split from keras.utils import to_categorical from tqdm import tqdm train = pd.read_csv('train.csv') # Reading the training images train_image = [] for i in tqdm(range(train.shape[0])):     img = image.load_img('Images/train/'+train['filename'][i], target_size=(28,28,1), grayscale=True) img = image.img_to_array(img) img = img/255     train_image.append(img) X = np.array(train_image) # Creating the target variable y=train['label'].values y = to_categorical(y) # Creating validation set X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42, test_size=0.2) # Define the model structure model = Sequential() model.add(Conv2D(32, kernel_size=(3, 3),activation='relu',input_shape=(28,28,1))) model.add(Conv2D(64, (3, 3), activation='relu')) model.add(MaxPooling2D(pool_size=(2, 2))) model.add(Dropout(0.25)) model.add(Flatten()) model.add(Dense(128, activation='relu')) model.add(Dropout(0.5)) model.add(Dense(10, activation='softmax')) # Compile the model # Training the model model.fit(X_train, y_train, epochs=10, validation_data=(X_test, y_test)) download = drive.CreateFile({'id': '1zHJR6yiI06ao-UAh_LXZQRIOzBO3sNDq'}) download.GetContentFile('Test_fCbTej3.csv') test_file = pd.read_csv('Test_fCbTej3.csv') test_image = [] for i in tqdm(range(test_file.shape[0])):     img = image.load_img('Images/test/'+test_file['filename'][i], target_size=(28,28,1), grayscale=True)     img = image.img_to_array(img)     img = img/255     test_image.append(img) test = np.array(test_image) prediction = model.predict_classes(test) download = drive.CreateFile({'id': '1nRz5bD7ReGrdinpdFcHVIEyjqtPGPyHx'}) download.GetContentFile('Sample_Submission_lxuyBuB.csv') sample = pd.read_csv('Sample_Submission_lxuyBuB.csv') sample['filename'] = test_file['filename'] sample['label'] = prediction sample.to_csv('sample.csv', header=True, index=False)

Submit this file on the practice problem page to get a pretty decent accuracy number. It’s a good start but there’s always scope for improvement. Keep playing around with the hyperparameter values and see if you can improve on our basic model.

Conclusion

Who said deep learning models required hours or days to train. My aim here was to showcase that you can come up with a  pretty decent deep learning model in double-quick time. You should pick up similar challenges and try to code them from your end as well. There’s nothing like learning by doing!

The top data scientists and analysts have these codes ready before a Hackathon even begins. They use these codes to make early submissions before diving into a detailed analysis. Once they have a benchmark solution, they start improving their model using different techniques.

Frequently Asked Questions Related

You're reading Build Your First Image Classification Model In Just 10 Minutes!

Zero Trust Network: Milestone In It Security Or Just Another Trust Model?

Zero Trust Network: Milestone In IT Security Or Just Another Trust Model? What Is Zero Trust Network?

There are several benefits of replacing the traditional system with Zero trust network. In the ecosystem of zero trust network model, it is assumed that we are working in an open environment which has endless threats and vulnerabilities. This makes sure that every data; incoming or outgoing is encrypted to avoid any mishappening!  It is a bit inconvenient for users as they have to log in for session as there is no provision of cookies that will keep them logged in. Also, the privileges of administrators are restricted. The admins are no longer permitted to access or use their power anytime, but during their working hours. Moreover, the systems are subdivided to be make sure they are ready to work with zero trust approach. These are divided separately in sections to prevent any other person from accessing the sensitive information.

Also Read : Same Old Cyber Security Failures

Best Practices for Zero Trust Network Model That You Must Start With

1. Identify the Most Sensitive Data You Have

You may have to invest a lot of time for this but it will be worth the effort as you’ll get an insight into the data and users who have access to it. You may now take the important steps and practices for securing it.

Well you won’t have to invest your time for this as you’ll be able to track this while carrying out the previous step! But document this to make sure you have a record of the people who have access to it and right to share the same!

3. Come Up with Regulations

As every user in the zero-trust network must know about the basics of what he is allowed to do and what not, this is crucial! Also, in case of any dispute you’ll have a guideline to settle it!

4. Keep Monitoring Continually

Everything will go in vain if your network is left unsupervised! So, make sure you have experts to track and supervise the activities in your network!

5. Get Rid of Toxic Data

In every organization, there is a dataset which is no longer useful but still the personnel’s involved find it hard to let go of it. If you too have the same, dispose it off immediately! It will not only free up some space for you but also make you less vulnerable to hackers.

Must read : Tips to Enhance Privacy on Chrome For Android

Yes, there are several things to upgrade, but switching to this from a traditional network will give a boost to your network security! So, start following zero trust approach for your network and stay shielded! Certainly, this will not make you cent percent secure but reduce the risks considerably! What do you think?

Quick Reaction:

About the author

Tweak Library Team

What Iphone Model Do I Have? Find Your Iphone Model

Many iPhone models look alike. Sure, if you are tech-savvy, you might know about the differences by seeing the camera style, antenna lines, dimensions, etc. But what about general folks! Well, if you are someone who has trouble knowing what iPhone model do you have or which one you are about to buy second hand, here’s an easier way to find out. Let me guide you through the process of identifying various iPhone models.

How Can I Find Out Which iPhone Model I Have?

Identify the iPhone model number using the steps below and confirm it with the list at the end of this post.

Use the Settings App

Turn on the iPhone and open the Settings app.

Tap on General and tap on About.

Tap on Model Number. You will see a number starting with ‘A.’ This is your iPhone’s model number.

Look Inside the Sim Slot

On newer iPhones (8 and later), the model number is etched inside the SIM slot. The tiny alphabets are tough to read. Make sure you have a flashlight or a second mobile phone with a torch.

Use the ejector to take out the SIM tray. Put it safely aside.

Now, point the flashlight inside the SIM slot. You will see the tiny text on the upper edge of the slot.

See the Back of the iPhone

On older iPhones (7 and earlier), the model number is written at the back of the device. It is easily visible.

Remove any case, cover, skin, or opaque protector from the back of your iPhone.

In the lower back, you will see the model number. It starts with the alphabet ‘A.’

Full List of iPhones with their Corresponding Model Number

Once you know the model number (after following one of the above methods), match that number with the list below to identify which iPhone you have.

iPhone 12 Pro Max

A2342 (United States)

A2410 (Canada, Japan)

A2412 (China mainland, Hong Kong, Macao)

A2411 (other countries and regions)

iPhone 12 Pro

A2341 (United States)

A2406 (Canada, Japan)

A2408 (China mainland, Hong Kong, Macao)

A2407 (other countries and regions)

iPhone 12

A2172 (United States)

A2402 (Canada, Japan)

A2404 (China Mainland, Hong Kong, Macao)

A2403 (other countries and regions)

iPhone 12 mini

A2176 (United States)

A2398 (Canada, Japan)

A2400 (China mainland)

A2399 (other countries and regions)

iPhone SE (2nd generation)

A2275 (Canada, United States)

A2298 (China mainland)

A2296 (other countries and regions)

iPhone 11 Pro Max

A2161 (Canada, United States)

A2220 (China mainland, Hong Kong, Macao)

A2218 (other countries and regions)

iPhone 11 Pro

A2160 (Canada, United States)

A2217 (China mainland, Hong Kong, Macao)

A2215 (other countries and regions)

iPhone 11

A2111 (Canada, United States)

A2223 (China mainland, Hong Kong, Macao)

A2221 (other countries and regions)

iPhone XS Max

A1921, A2101

A2102 (Japan)

A2103

A2104 (China mainland)

iPhone XS

A1920, A2097

A2098 (Japan)

A2099

A2100 (China mainland)

iPhone XR

A1984, A2105

A2106 (Japan)

A2107

A2108 (China mainland)

iPhone X

A1865, A1901

A1902 (Japan)

iPhone 8 Plus

A1864, A1897

A1898 (Japan)

iPhone 8

A1863, A1905

A1906 (Japan)

iPhone 7 Plus

A1661, A1784

A1785 (Japan)

iPhone 7

A1660, A1778

A1779 (Japan)

iPhone 6s Plus

A1634, A1687, A1699

iPhone 6s

A1633, A1688, A1700

iPhone SE (1st generation)

A1723, A1662, A1724

iPhone 6 Plus

A1522, A1524, A1593

iPhone 6

A1549, A1586, A1589

iPhone 5s

A1453, A1457, A1518, A1528, A1530, A1533

iPhone 5c

A1456, A1507, A1516, A1529, A1532

iPhone 5

A1428, A1429, A1442

iPhone 4s

A1431, A1387

iPhone 4

A1349, A1332

iPhone 3GS

A1325, A1303

iPhone 3G

A1324, A1241

iPhone (The 1st iPhone)

A1203

Signing Off

This is how you can know what iPhone do you have. Once you know this, it becomes easier to learn more about the device and its specific features.

READ NEXT:

Author Profile

Ankur

I have been an Apple user for over seven years now. At iGeeksBlog, I love creating how-tos and troubleshooting guides that help people do more with their iPhone, iPad, Mac, AirPods, and Apple Watch. In my free time, I like to watch stand up comedy videos, tech documentaries, news debates, and political speeches.

How To Play A Sound Every Few Minutes On Windows 10

A professional worth his/her salt knows how easy it is to lose track of time when caught up with work. An alarm, then, is no less than a friend reminding us of other things that also need our attention. Most of us can set an alarm on our phone without even looking at it. But it is a different story with the desktop.

Windows 10 clock app is as minimalist in design as it is in terms of features – other than an alarm and snooze, there’s not much else. That is why most people set reminders on phones even when they’re on their computer, which isn’t the most efficient way of going about things. 

Well, here’s a way you can play a sound – a beep, a chime, a snippet of a song – every few minutes and keep track of time in a better way. 

Step #1: Get cmdmp3

First up, you have to download cmdmp3. This is a command-line mp3 player which, once executed, will play the sound that you wish. Here is the download link:

Download: cmdmp3new.zip

Once you download it, extract its contents using a file archiver/unzipper. Take note of where you extracted the contents of the zip file. 

Step #2: Create a batch file in the “cmdmp3” folder

Now, it is time to create a batch file that uses cmdmp3 and the sound file of your choosing to go along with it.  First up, copy and paste your sound file to the “cmdmp3new” folder. 

Type: cmdmp3 *file-name-of-your-sound-file*

In place of *sound file name*, type in the name of your sound file (including its extension). See the example below:

You can use .MP3 and .WAV files, but some other extensions may also work. 

Now, your batch file should appear inside the cmdmp3 folder. Meaning, both your .BAT file and the sound file should be in the cmdmp3 folder where you also have the cmdmp3 file (cmdmp3.exe).

Alternative: Create a batch file anywhere

Alternatively, if you want to keep your batch file in some other location (say, on the desktop, for easy access), there’s a slight modification to the method above. Basically, you’ll have to enter the entire address of both the chúng tôi and the sound file in the notepad.

Open Notepad as shown before. This time, we will include the full location of chúng tôi and the sound file. Here’s how you can easily copy the files’ path location:

Then paste it into the notepad file as well. Make sure there is a space between the cmdmp3 address and the sound file’s address. 

Step #3: Create a task in Task Scheduler

Now, we have to make sure this batch file that plays the sound runs automatically every few minutes. This we will do through Task Scheduler. 

Under the General tab, give the task a name.

The “Begin the task” option will be set to On a schedule by default. Keep it that way. Make sure the start date is set to today’s date, and the time 12:00:00. This will make sure the task runs daily. 

Now, your audio file will sound a reminder every few minutes, depending on the time period that you chose. Do remember that if you want to create a batch file other than in the cmdmp3 folder, you have to pick the #2 Alternative method. 

How To Build Your Own Cotton

You don’t need to wait for a carnival to satisfy your craving for cotton candy. Instead, build this portable, pocket-size machine to turn granulated sugar into an airy treat.

A DIY cotton-candy machine consists of a small metal container, repurposed lighter parts to provide heat, and a switch-controlled motor to set everything spinning. Slowly pour granulated sugar into the container, and flames from the lighters will melt it. As the motor spins, the liquid sugar will fly out through little holes in the container’s sides, forming thin strands. A paper cylinder placed around the machine will capture them. Once they’ve built up, simply swirl a chopstick around the perimeter to gather the candy and taste your sweet success.

This article originally appeared in the September 2024 issue of Popular Science, under the title “Build Your Own Cotton-Candy Machine.”

Stats

Time: 2 hours

Cost: $26

Difficulty: Medium

Tools

Push pin

Power drill

Soldering iron

Materials

Long-nosed lighter

Torch lighter

Wire

Two-part epoxy

Superglue

Metal stand-off with a screw, washer, and nuts

A small cosmetic aluminum container (found in drugstores) or a metal drink cap

Small project box

DC motor

AA-battery holder

Clay epoxy

Paper, tape, rubber band, and a chopstick

Instructions

To build a system for heating the sugar, first open both lighters. Harvest the large fuel tank, igniter, and hose from the long-nosed lighter and the torch head from the torch lighter.

Use the long fuel hose to connect the fuel tank to the torch head.

For an ignition line, wrap a short length of wire around the metal base of the long- nosed lighter’s igniter and seal it with epoxy.

Push the igniter’s new wire through the torch head— where the torch lighter’s wire previously was. This is the main ignition line.

Connect the main ignition line to the brass part of the torch head. Seal with superglue.

Next, set up the spinning chamber. Epoxy the metal standoff to the shaft of the motor. (When joining two parts together with epoxy, sanding both sides will yield a stronger bond.)

With the push pin, punch holes all the way around the sides of the aluminum container, or drill tiny holes in the metal drink cap. Find the center of the container and drill through it. Add the screws, washers, and bolts to it, and screw it in place on the motor’s standoff.

Solder the battery pack’s terminals to the motor. Since the screw tightens clockwise, make the motor spin counterclockwise to prevent it from unscrewing.

To prepare the project box, plan where you will be placing the fuel valve, igniter, torch head, and spinning chamber. Mark each spot with a marker, and then drill the holes. You can use the photos as a guide.

Inside the cotton-candy machine Sophie Bushwick

Epoxy the motor in place in the box. Glue the battery pack to the outer side of the box. Seal the igniter in place—the end should stick out of the box—with clay epoxy.

Before sealing the torch system in place with the clay epoxy, measure the torch head and aim it at an angle so the flame will touch the near edge of the metal container.

To operate the cotton-candy machine, tape paper into a cylinder that fits around it. Then switch on the motor, squeeze the fuel valve (and hold it in position with a rubber band), and spark the igniter. Let the machine heat for 10 seconds, then place the paper cylinder around it and slowly add the sugar. Collect the candy with a chopstick.

Warning: Take care handling lighters and fuel. The sugar is molten when it comes out, so keep your hands out of the way of hot flying sugar. Also keep hands and paper clear of the open flame—or you might end up making jerky instead of candy.

Windows 10 Insider Preview Build History Tracker

UPDATED 11/18/21: As part of the Windows 10 development, Microsoft has the Windows Insider Program, a testing program that allows developers, businesses, and enthusiasts to access early builds of the operating system to test new features and changes and validate software and drivers before deployments. In this guide, you can track the history of every Windows 10 Insider Preview build for PC, including information about their release date, changelog summary, build number, and more.

The program includes three readiness levels (also known as channels). The Dev Channel (formerly Fast ring) is for highly technical users since builds are rough and contain code that is still under heavy development. Usually, these builds are not stable, you’ll come across errors and problems, and you may have to use workarounds to fix problems that may affect your experience.

The Beta Channel (formerly Slow ring) is for anyone who wants to test upcoming features or needs to validate apps and deployments. Previews in this channel are stable, but you may still run into issues. Also, features included in this channel are expected to ship with the next release of Windows 10.

If you have a device enrolled in any of these channels, you can use this guide to track the history of every released Windows 10 Insider Preview build.

Windows 10 21H2 Insider Preview build history

Windows 10 21H2 is the eleventh feature update expected to arrive in the fall of 2023. Some of the features and changes planned for this new version are already available through the Dev Channel.

Below you’ll find the history of the Windows Insider Preview builds released in the Dev Channel. The latest version available for testers is Windows 10 build 19044.1381.

Release dateBuild numberChannelPlatformNotes

Fixes.

Fixes.

Fixes.

Fixes.

WSL features.

Improvements.

Although Microsoft hasn’t shared any official details, it’s expected that the company will skip the 21H1 update schedule and release version 21H2 during the fall of 2023 instead.

Windows 10 21H1 Insider Preview build history

Windows 10 21H1 is the eleventh feature update expected to arrive in the spring of 2023. This refresh will be considered a minor release with a small set of features and several fixes that will be delivered as a quick enablement package for version 2004 or 20H2.

Below you’ll find the history of the Windows Insider Preview builds released in the Beta Channel. The latest version available for testers is Windows 10 build 19043.1381.

Check out all the features and changes coming to Windows 10 21H1.

Windows 10 20H2 (October 2023 Update) is the tenth feature update expected to release in October 2023. Below you’ll find the history of the Windows Insider Preview builds released in the Beta Channel (formerly Slow ring). The latest version available for testers is Windows 10 build 19042.1023.

Check out all the new features and changes expected to arrive in Windows 10 20H2.

Windows 10 version 2004 Insider Preview build history

Windows 10 version 2004 (May 2023 Update) — also known as the 19H1 update — is the ninth feature update released on May 21, 2023. Below you’ll find the history of the Windows Insider Preview builds released in the Fast, Slow, and Release Preview rings. The latest version available for testers is Windows 10 build 19041.264.

Check out all the features and changes coming to Windows 10 version 2004 (20H1).

Release dateBuild numberRingPlatformNotes

KB4556803

KB4558244

KB4558244

56April 17, 202319041.207Release Preview ringPCMinor update.

KB4550936

KB4552455

KB4552455

KB4541738

KB4540409

KB4539080

KB4535550

Quick searches for Search Home.

Notepad old version restored.

Version 2004.

Almost RTM.

44November 20, 2023 19025.1051Slow ringPCMinor update.

43November 19, 2023 19028Fast ringPCMinor update.

Search algorithm update.

41November 12, 2023 19023Fast ringPCMinor update.

40November 11, 2023 19013.1122Slow ringPCMinor update.

39November 5, 2023 19018Fast ringPCMinor update.

RDP fix.

KB4528332.

Updated Kaomoji.

No new features.

No new features.

No new features.

PIN on Safe mode.

Narrator improvements.

Windows Update with optional updates.

Vibranium codename reference.

Cortana resize and drag.

New tablet mode experience.

KB4517787 (test).

New Feedback app.

Notepad now Microsoft Store app.

File Explorer drops Homegroup options context menu.

Reset this PC with Cloud download option.

Windows Defender renamed to Microsoft Defender.

Korean IME update.

Passwordless settings.

Sync settings engine upgrade.

Fixes.

Windows Ink Workspace update.

15June 5, 2023 18912Fast ringPCMinor update.

YourPhone new features.

Drive type on Task Manager.

Fixes Your Phone app.

File Explorer cloud search.

Kills Friendly Dates.

SwiftKey languages.

Notepad unsaved recovery.

No new features.

No new features.

No new features.

No new features.

Windows 10 version 1909 Insider Preview build history

Codenamed 19H2, Windows 10 version 1909 (November 2023 Update) was the eighth next major update released on November 12, 2023, as a cumulative update with new features and changes. Below you’ll find the history of the Windows Insider Preview builds released in the Fast, Slow, and Release Preview rings. The latest version available for testers is Windows 10 build 18362.10024 (Slow ring). If you’re in the Release Preview ring, the version number is Windows 10 build 18363.418.

Release dateBuild numberRingPlatformNotes

Moves build 18362.10024 to 18363.418

Enables all features for version 1909.

Fixes.

Fixes.

Fixes.

Fixes.

Not final version.

Not final version.

All features enabled.

Two fixes.

Not final version.

Same features as build 18362.10013, but disabled.

3July 17, 202318362.10006SlowPCMinor update.

3rd-party assistants Lock screen.

No new features.

No new features.

Windows 10 version 1903 Insider Preview build history

Codenamed 19H1, Windows 10 version 1903 (May 2023 Update) was the seventh major update released on May 21, 2023. Below you’ll find the history of the Windows Insider Preview builds released in the Fast, Slow, and Release Preview rings. The latest version available to Insiders in the Fast ring is Windows 10 build 18362 and build 18362.30 in the Slow ring. On April 8, 2023, Microsoft is making available Windows 10 build 18362.30 as the final version for the May 2023 Update in the Release Preview ring.

Check out all the features and changes coming to Windows 10 19H1.

Release dateBuild numberRingPlatformNotes

42April 8, 2023 18362.30Release PreviewPCFinal version

Fast RingPCKB4497464

40April 3, 2023 18356.21Slow ringPCKB4496796

Bug fixes.

Bug fixes.

Bug fixes.

36March 19, 2023 18356.16Slow ringPCKB4494123

Bug fixes.

Quality update.

33March 14, 2023 18351.26Slow ringPCKB4494195

32March 13, 2023 18351.8Slow ringPCKB4492310

Your Phone mirroring.

30March 11, 2023 18351.7Slow ringPCKB4492310

Windows Sandbox changes.

Bug fixes.

Bug fixes.

Bug fixes.

Bug fixes.

Bug fixes.

Bug fixes.

Bug fixes.

Mixed Reality with Win32 app support.

Light theme improvements.

Insider settings updated.

Disable acrylic option for Sign-in screen.

Cortana is off during clean install.

Symbols on emoji panel.

Windows Update changes.

Emoji 12.

Indic Phonetic keyboards.

Touch keyboard improvements.

Automatic troubleshooter.

Ebrima font.

No new features.

No new features.

Sign-in screen acrylic effect.

No new features.

No new features.

No new features.

No new features.

Windows 10 version 1809 Insider Preview build history

Dubbed Redstone 5, Windows 10 version 1809 (October 2023 Update) was the sixth major feature update released on November 13, 2023. Below you’ll find the history of the Windows Insider Preview builds released in the Fast, Slow, and Release Preview rings. The latest version available to Insiders in the Fast ring is Windows 10 build 17763 and build 17763 in the Slow ring.

Release dateBuild numberRingPlatform

October 30, 202317763.107Slow ring and Release Preview PC

October 16, 202317763.104Slow ring and Release Preview PC

October 9, 202317763.17Slow ring and Release Preview PC

September 20, 202317763Slow ringPC

September 18, 202317763Fast ringPC

September 18, 202317758.1004Slow ringPC

September 15, 202317760Fast ringPC

September 14, 202317758Slow ringPC

September 11, 202317758Fast ringPC

September 11, 202317754Slow ringPC

September 7, 202317755Fast ringPC

September 5, 202317754Fast ringPC

August 31, 202317751Fast ringPC

August 30, 202317744Slow ringPC

August 29, 202317738Slow ringPC

August 24, 202317746Fast ringPC

August 21, 202317744Fast ringPC

August 17, 202317741Fast ringPC

August 14, 202317738Fast ringPC

August 10, 202317735Fast ringPC

August 8, 202317733Fast ringPC

August 3, 202317730Fast ringPC

July 31, 202317728Fast ringPC

July 26, 202317713.1002Slow ringPC

July 25, 202317723Fast ringPC

July 23, 202317713.1002Fast ringPC

July 11, 202317713Fast ringPC

July 7, 202317711Fast and Skip AheadPC

July 2, 202317692Slow ringPC

June 27, 202317704Fast and Skip AheadPC

June 14, 202317692Fast and Skip AheadPC

June 6, 202317686Fast and Skip AheadPC

May 31, 202317682Fast and Skip AheadPC

May 24, 202317677Fast and Skip AheadPC

May 16, 202317672Fast and Skip AheadPC

May 9, 202317666Fast and Skip AheadPC

May 3, 202317661Fast and Skip AheadPC

April 25, 202317655Skip AheadPC

April 20, 202317650Skip AheadPC

April 12, 202317643Skip AheadPC

April 4, 202317639Skip AheadPC

March 30, 202317634Skip AheadPC

March 21, 202317627Skip AheadPC

March 16, 202317623Skip AheadPC

March 7, 202317618Skip AheadPC

February 14, 202317604Skip AheadPC

Windows 10 version 1803 Insider Preview build history

Version 1803 (April 2023 Update) was the fifth feature update of Windows 10 released on April 10, 2023. Below you’ll find the history of the Windows Insider Preview builds released in the Fast, Slow, and Release Preview rings for the Redstone 4 update development. The latest version available to Insiders is Windows 10 build 17134.

Release dateBuild numberRingPlatform

April 20, 202317134Release PreviewPC

April 20, 202317134Slow RingPC

April 16, 202317134Fast RingPC

April 5, 202317133Release PreviewPC

March 30, 202317133SlowPC

March 27, 202317133FastPC

March 23, 202317128FastPC

March 23, 202317127SlowPC

March 20, 202317127FastPC

March 16, 202317123FastPC

March 16, 202317120SlowPC

March 13, 202317120FastPC

March 9, 202317115SlowPC

March 6, 202317115FastPC

March 2, 202317112FastPC

February 27, 202317110FastPC

February 23, 202317107FastPC

February 14, 202317101FastPC

February 7, 202317093Fast and Skip AheadPC

January 24, 202317083Fast and Skip AheadPC

January 19, 202317074.1002Slow PC

January 19, 202317074.1002Fast and Skip AheadPC

January 12, 202317074Fast and Skip AheadPC

December 19, 202317063Fast and Skip AheadPC

November 22, 202317046Fast and Skip AheadPC

November 16, 202317040Fast and Skip AheadPC

November 8, 202317035Fast and Skip AheadPC

November 1, 202317025SlowPC

October 25, 202317025Fast and Skip AheadPC

October 13, 202317017Fast and Skip AheadPC

September 27, 202317004Fast (Skip Ahead)PC

September 13, 202316362Fast (Skip Ahead)PC

August 31, 202316353Fast (Skip Ahead)PC

July 26, 202316251Fast (Skip Ahead)PC

Windows 10 version 1709 Insider Preview build history

Windows 10 version 1709 (Fall Creators Update), internally known as Restore Store 3, was the fourth feature update. You’ll find this history of all previews available for this release before it became available to everyone in the table below. The currently available version is Windows 10 build 16299.15 for PC and Windows 10 Mobile build 15254.1 (Windows 10 Mobile Fall Creators Update candidate) for phones.

Release dateBuild numberRingPlatform

October 19, 202315254.1Release PreviewMobile

October 11, 202315254.1Fast/SlowMobile

October 10, 202316299.15Release PreviewPC

October 4, 202316299.15SlowPC

October 2, 202316299.15FastPC

September 28, 202316299SlowPC

September 26, 202316299FastPC

September 26, 202316296SlowPC

September 22, 202316296FastPC

September 22, 202315252SlowMobile

September 22, 202316291SlowPC

September 22, 202316296FastPC

September 20, 202316294FastPC

September 19, 202316291FastPC

September 15, 202315252FastMobile

September 15, 202316288SlowPC

September 14, 202315245SlowMobile

September 12, 202316288FastPC

September 12, 202315250FastMobile

September 1, 202316278SlowPC

September 1, 202316281FastPC

August 29, 202316278FastPC

August 25, 202315245FastMobile

August 25, 202316275FastPC

August 24, 202315240SlowMobile

August 23, 202316273FastPC

August 9, 202315240FastMobile

August 2, 202316257FastPC

August 2, 202315237Fast Mobile

July 26, 202316251Fast (rs3_release)PC

July 26, 202315235FastMobile

July 13, 202316241FastPC

July 13, 202315230FastMobile

July 7, 202316237FastPC

July 7, 202316232SlowPC

July 6, 202316232.1004FastPC

June 28, 202316232FastPC

June 28, 202315228FastMobile

June 21, 202316226FastPC

June 21, 202315226FastMobile

June 21, 202315223SlowMobile

June 13, 202315223FastMobile

June 8, 202316215FastPC

June 8, 202315222FastMobile

May 17, 202316199FastPC

May 17, 202315215FastMobile

May 11, 202316193FastPC

May 11, 202315213FastMobile

May 4, 202316188FastPC

April 28, 202315208FastMobile

April 28, 202316184FastPC

April 24, 202315207FastMobile

April 19, 202316179FastPC

April 19, 202315205FastMobile

April 14, 202316176FastPC

April 14, 202315204FastMobile

April 7, 202316170FastPC

Windows 10 Creators Update (Redstone 2)

April 25, 202315063.251Release Preview, Slow Mobile

March 29, 202315063SlowPC

March 29, 202315063.2SlowMobile

March 28, 202315063.2FastPC

March 20, 202315063SlowPC

March 20, 202315063FastPC, Mobile

March 17, 202315061FastPC

March 16, 202315060FastPC

March 14, 202315058FastPC

March 10, 202315055FastPC, Mobile

March 10, 202315051SlowMobile

March 8, 202315051FastMobile

March 8, 202315048SlowPC

March 3, 202315048FastPC

March 3, 202315047FastMobile

March 2, 202315043SlowMobile

March 1, 202315042SlowPC

February 28, 202315046FastPC

February 24, 202315043FastMobile

February 24, 202315042FastPC

February 8, 202315031FastPC, Mobile

February 3, 202315025SlowPC

February 1, 202315025FastPC, Mobile

January 27, 202315019FastPC

January 20, 202315014FastPC, Mobile

January 12, 202315007FastPC, Mobile

January 9, 202315002FastPC

December 8, 202414986FastPC

December 1, 202414977FastMobile

November 17, 202414971FastPC

November 16, 202414965SlowPC, Mobile

November 9, 202414965FastPC, Mobile

November 3, 202414959FastPC, Mobile

October 25, 202414955FastPC, Mobile

October 19, 202414951FastPC, Mobile

October 13, 202414946FastPC, Mobile

October 7, 202414942FastPC, Mobile

September 28, 202414936FastPC, Mobile

September 14, 202414926FastPC

August 31, 202414915FastPC, Mobile

August 17, 202414905FastPC, Mobile

August 11, 202414901FastPC

If you don’t know the version of Windows 10 your PC is running, you can use this guide to determine exactly the version of the operating system installed on your device.

Join Windows Insider Program

On Windows 10, using the Settings app, you can join your device into any of the three testing changes to get early access to new features and enhancements.

To join the Windows Insider Program, use these steps:

Open Settings.

Sign in with your Microsoft account.

Under the “Pick your Insider settings” section, select the previews you want to test, including:

Dev Channel — Includes new features as soon as they’re ready, but builds are not stable.

Beta Channel — Includes new features and enhancements expected for the next version of Windows 10. The previews are more stable, but you may still run into issues.

Release Preview Channel — Includes the final version of Windows 10, minus some quality updates.

Open Settings after the reboot.

Once you complete the steps, the latest preview build of Windows 10 available in the channel you selected will download and install on your computer.

Switch Windows Insider Program channel

You can always switch between channels to test different builds, but you can’t roll back to the current version of Windows 10 without reinstallation.

To switch Insider channels, use these steps:

Open Settings.

Under the “Pick your Insider settings” section, select the active Insider channel – for example, Dev Channel.

Select a different Insider channel of Windows 10.

Restart your device.

Open Settings after the reboot.

After you complete the steps, the new build from the active channel you selected will download and install on the device.

Stop Windows Insider Program builds

Although you can’t roll back to the current release of Windows 10 (unless you perform a clean installation), you can use an option to opt-out of the program, which will continue to download Insider builds. However, once the final version is released, the device will gracefully be removed from the program.

Stop Windows Insider Preview builds

To stop getting Insider builds of Windows 10, use these steps:

Open Settings.

Under the “Stop getting preview builds” section, turn off the toggle switch for the setting.

Once you complete the steps, the device will no longer receive Insider builds once the final version is installed.

Originally published in October 2023, information has been updated in Novmeber 2023, and it’ll be kept updated as new releases become available.

Update the detailed information about Build Your First Image Classification Model In Just 10 Minutes! on the Moimoishop.com website. We hope the article's content will meet your needs, and we will regularly update the information to provide you with the fastest and most accurate information. Have a great day!