CodeChem

codechem

The year was 2007. There were two guys who just moved to San Francisco. Both were unemployed, so they faced the trouble of paying rent. Suddenly, they realized that all the hotel rooms in San Francisco were reserved because the local industrial design conference at the time attracted a large number of visitors. And this is how it all started... A minimum viable product was born. They came up with the idea of providing accommodation for people coming to town. They took a few pictures of their apartment, launched a basic website, and promoted it to people who were coming to San Francisco for a design conference. And this is how Brian and Joe built AirBnB! Brian and Joe's concept is what we call an MVP Development nowadays. WHAT IS AN MVP? MVP is a version of your product you want to launch with and features. Besides that, those futures should also cover 3 other areas: minimal functional = no bugs, no crashes, no disappointment. Reliability = simple UI and legibility. Usability = unique, lovable, friendly, and usable. User Experience This MVP Pyramid Model from Jussi Pasanen perfectly illustrates the issue: "Build a slice across, instead of one layer at a time". Be sure that less functionality doesn't mean poor quality. In fact, in a competitive digital world, the product you build to achieve user traction needs to be more than just a few lines of code well kept into a repository. It needs to fit into the everyday environment, create emotion, solve real problems, and tell a story while it does all that. By going with an MVP first, you can drastically increase your chances of success. It allows you to fit into the market in a short time with a small budget and satisfy early adopters. It is very important to note that an MVP is a process of collecting feedback. It’s not a finished product. In other words, once the minimum viable product is launched, you await valuable feedback for future product developments. Тhat will allow you to continue improving and developing your MVP. One day, your MVP will become a fully-fledged product. One of the places for collecting feedback for your MVPs is the place where early adopters like to spend their free time — on ProductHunt. We all know that launching a new product is a complex process and can be tricky, especially for startups. Most failures happen in a scenario where there is no research on the needs and limits of the market, adequate cash flow, or if the product is not user-friendly. Now that we’ve established what an MVP is, let’s figure out how to build an MVP right. So, you should follow these 5 steps: Analyze the market and your competitors, Figure out what problem you’re solving, and for whom, List all necessary features and prioritize them, Develop your MVP, Launch! 🚀 Step 1. Analyze the market and your competitors It may happen that the idea doesn't fit the needs of the market. Before you start building an MVP, make sure it fulfills the needs of your target users. Cover the most suitable market for the start. In our experience, it is not optimal to spend most of the resources only on market research especially if it is a new product. If there is already a product similar to the one you want to launch on the market, it is important to do a competitor analysis. It'll help you make your product more unique. You can be inspired by the competition and tweak the product in order to gain a competitive advantage. Another crucial step before you start building the MVP is to get early feedback from experts. They can give you valuable and wise advice and guide you along the way to completing your future MVP! Step 2. Figure out what problem you’re solving, and for whom: The first thing of successful MVP development is your idea. Start by asking yourself: validating Who will my product? use What (time, money) does my product offer? value What do I solve? problem What is the ? core functionality What determinates ? success and failure can I deliver it? When When and how much will it bring? revenue Step 3. List all necessary features and prioritize them Don't forget that the MVP must be low-cost, practical, and delivered in the short time scope previously defined. As soon as you’ve defined all the steps, you can start creating a list of particular features for each stage. Then, you need to prioritize them. For better understanding, we can define a user flow for our core features. One of the most important questions is what will make your users intrigued and happy to see. The core features of your product, The should-have features, The nice-to-have features. When you’ve prioritized all the features, you may define their scope for the first version of the product and start to build the MVP. By creating a prototype you can have a better grasp of how your future product would look like. Consider prototyping as an MVP to build an MVP: not a fully functional version, but a version to help visualize the look and feel of the product. In our experience, an iterative approach would be the best solution to move forward to our common goal, which is the MVP itself. Step 4. Develop your MVP (Coding Time) Since you researched, deliberated, and made many hard decisions, it’s finally time to translate theory into practice! You assemble a team to do the . UX, UI, Code, DevOps, and Tests Keep in mind that an MVP is a product that is born based on . And that process requires experience. You should develop the MVP with speed and precision. It's advisable to have a team of agile software developers who have expertise in specific technologies, implementation, and testing that in the end will save you resources and successfully reach the goal (MVP). PROCESS Step 5. Launch! 🚀 Finally, it’s time to get your product out in the world. That means you allow the early adopters to try your product. You need to pay attention to the feedback from them, as this will be the cornerstone for further development. There are several approaches to go-to-market strategy of an MVP launch. Soft launch The soft launch requires releasing to a limited half of the target audience. The main idea of soft product launching is to reduce risks with product release. It is based on a few key points: Collecting users' opinions helps to improve the usability and user journey. It also allows prioritizing features for the further development. Receive the users’ feedback. Understand the user willingness to pay. Test the monetization options. Hard launch Hard launch is not the best option for launching your MVP. It's applicable to the mature products because the release is with huge marketing activities. The hard launch is a good option if: You have strong infrastructure. You have a large and predictable audience. You can successfully anticipate the market reaction. Dark launch The dark launch is based on the process of continuous delivery. In other words, it’s a regular feature update of your MVP. This approach allows you to track the user behavior in real-time. It’s a perfect way to decide, whether a new update is efficient enough to scale to all users. Because at the end of the day, a successful product is always a result of planning and thorough analysis. So, is it possible to build an MVP in 3 months? 🚀 It sure is! MVP in itself is considered to be a time-efficient solution for a number of reasons, including the focus only on the essential features and the opportunity to initiate the feedback loop early on. In some cases, the development may be restricted to a month or two – from our own experience, however, three months is the optimal timeframe for building your application just enough for your users to get a feel of what it's all about. You might be very excited when coming up with an idea for software that can change people’s lives and make it easier and entertaining. At CodeChem, we also get excited when we get to take part in creating something unique, lovable, friendly, and useful. Let’s combine your ideas with our capabilities and expertise to develop an awesome product. If you’re ready, check out our process and drop us a line.

How to Build a Successful MVP in 3 Months 🚀

For us, developers, starting a new web project can be a tedious job. It raises questions like: What technology should I use for the front-end? What technology should I use for the back-end? What database is the best? Because nowadays all Javascript technologies like , and are very popular for building Web applications, so we can get an answer for the first question very fast. React Angular Vue But what about the back-end? Should I use or ? Is it better to use a or database? Well, Strapi has the answer to all these questions. NodeJS .NET Core relational non-relational What is Strapi? Strapi is an open-source Headless CMS that gives developers the freedom to choose their favorite tools and frameworks. With all the plugins and features Strapi gives the developers the ability for customization and extensibility. Strapi is also very Developer-Friendly by providing an API that can be easily accessed either via REST or GraphQL endpoint. In this article, we are going to build a simple web application using Strapi and Angular. First, we are going to set up and build our . API Install Strapi Requirements: 12.x and npm 6.x Node.js Create a new project npx create-strapi-app blog_api --quickstart If we want to customize our application to use the specific database we need to run the command without argument. By default, Strapi is using . --quickstart sqlite Once the setup from the command above is finished Strapi will automatically run (NOTE: when manually start the project run the command ) and we can navigate to our admin panel on the following link: . When you navigate you will able to see the registration form. strapi develop http://localhost:1337/admin When we finish with registering our first user, we can start building our API. First, what we need to do for our Blog Application is to define the models that we will have. We will define three models: , , and . Page Post Content To create a new Model navigate to Content Type Builder. Our model will have: Content Title - type Text Value - type RichText IsPublished - type boolean CoverImage - type Media Relation to (Content belong to many ) Post Posts Relation to (Content belong to many ) Page Pages model will have: Page Name - type Text Relation to ( has many ) Content Page Contents Relation to ( has many and belongs to many ) Post Page Posts and model will have: Post IsDeleted - type boolean Relation to ( has many and belongs to many ) Page Post Pages Relation to ( has many ) Contents Post Contents As soon as we define our models we are ready to create some pages, contents, and posts. We can simply do that by navigating to each model and click Add new [name-of-the-model] Now, when we have models and data into our database we need to give access to our visitors of blog application. To do that we need to navigate to tab in the menu. We can see there are by default two types of roles: and . We navigate to role and select: Roles and Permissions Public Authorized Public And that's it. Our API is ready. Now we only need to make our web application. Angular Application For creating our Angular application we will use Angular CLI. Install the Angular CLI with the following command: npm install -g @angular/cli Run the following commands to create and run a new angular app: ng blog-web 
cd blog-web 
ng serve new If you navigate to you will able to see the new app. http://localhost:4200/ Now, we can start with styling our application and access data from our API. First, we will create services and API calls to get our data from Strapi. Navigate to folder and run the following commands: src mkdir services
cd services
ng generate service page
ng generate service post
ng generate service content Angular CLI will create these services so we are ready for coding. In we will put our API URL. environment.ts Navigate to page service and insert the following code: page-service.ts We created two methods: one for getting all pages and one for getting page by id. We will make the same for and services. post content NOTE: Before using we need to register into HttpClient app-module.ts Go to app-module.ts Import the from , HttpClientModule @angular/common/http { HttpClientModule } import from '@angular/common/http' Add it to the array. @NgModule.imports imports:[HttpClientModule,  ...] post-service.ts content-service.ts Next, we will create that will contain style and functionality for posts and we will use for displaying our landing page. Navigate to folder and create a new folder called components. Here, we will store all components that we use in our blog application. With the following command we can generate a new component: post-component app-component app ng generate component post Insert the following code into the post component. post.component.html post.component.scss post.component.ts Because we are using bootstrap classes we need to include bootstrap into our project as well. We can do that by adding the following into index.html <link rel= href= > "stylesheet" "https://stackpath.bootstrapcdn.com/bootswatch/4.3.1/cosmo/bootstrap.min.css" And we are almost done. The only thing that left is to modify and our blog is ready for use. app-component app.component.html app.component.scss app.component.ts Congratulations, we successfully built a Blog application. Conclusion Feel free to continue working on your blog. You can try various scenarios navigation, styling e.t.c. Play with models into Strapi and API calls from your Angular application. The source code of this tutorial is available on . I hope you enjoyed through this adventure. Github Also published on https://blog.codechem.com/the-power-of-strapi-building-web-application-has-never-been-easier

Introduction to Strapi : Headless CMS for Building Web Applications

Data persistence plays a big role in almost every application there is. With the introduction of .NET Core and EntityFramework Core a new possibility arose when it comes to data persistence. Now, it is possible to use EntityFramework Core to generate a local storage for our Xamarin Applications. In this article we will explain in a step-by-step tutorial how to set up a Xamarin Forms App with a SQLite local storage, and we are going to use EntityFramework Core to generate a Code-First Database and apply that migration to our app. I assume that you have basic knowledge and have installed the .NET Core SDK, .NET Core tools EntityFramework Core, and all other tools needed to develop a Xamarin Forms app. If not, I suggest you get familiar with them first. I am using Rider as my IDE. Note: Resources: https://docs.microsoft.com/en-us/ef/core/ https://docs.microsoft.com/en-us/ef/core/miscellaneous/cli/dotnet https://docs.microsoft.com/en-us/ef/ef6/modeling/code-first/migrations/ Setting up the project Let us go ahead and create a new Xamarin Forms App. Nothing fancy here, just do the usual project creation from the wizard. My project will be called XamarinEFCore. install and in every project, the traditional and the class library. Next: Microsoft.EntityFrameworkCore Microsoft.EntityFrameworkCore.Sqlite Then, create a new class library which will contain all the class entities we will store in our database. when you select the target framework for the class library, it should be .NET Standard 2.0 and not .NETCore x.x .Also install the here as well. Be careful Microsoft.EntityFrameworkCore You will need it for data annotations. Next, we will create our first entity. It is a pretty basic one just for demo purposes. public {
   [DatabaseGenerated(DatabaseGeneratedOption.Identity)]
   public int Id { get; set; }
   public string RandomGuid { get; set; }
   public DateTime TimeAdded { get; set; }
} class ListItem The data annotation on top of the Id property serves to generate an auto-increment primary key. If you plan to handle unique keys yourself, then you can use the annotation. [Key] Setting up the DatabaseContext The next step is to reference the entities class project in the traditional class library. This is needed so when we setup the context we can define the data sets. Then, we create a folder called that will contain our context class. Our context class will look like this: DatabaseContext public {
    public DbSet<ListItem> Items { get; set; }
    private string DatabaseName = ;

    protected override OnConfiguring(DbContextOptionsBuilder optionsBuilder)
    { databasePath; (Device.RuntimePlatform)
        { Device.iOS:
                databasePath = Path.Combine(Environment.GetFolderPath(Environment.SpecialFolder.MyDocuments), , , DatabaseName); ; Device.Android:
                databasePath = Path.Combine(Environment.GetFolderPath(Environment.SpecialFolder.Personal), DatabaseName); ; : NotImplementedException( );
        }
        optionsBuilder.UseSqlite($ );
    }
} : class ApplicationDbContext DbContext const "myItems.db" void String switch case ".." "Library" break case break default throw new "Platform not supported" "Filename={databasePath}" We add a property for each entity that we will store in the database. Because every platform has a different storage location, we set the databasePath accordingly based on the device platform. DbSet< Class > Next, we must do some work in the iOS project to make this work. In the add this line before AppDelegate.cs Xamarin.Forms.Init(); SQLitePCL.Batteries_V2.Init (); On the file add these two lines on top of the namespace: Main.cs [assembly: Preserve( (System.Linq.Queryable), AllMembers = )]
[assembly: Preserve( (System.DateTime), AllMembers = )]
[assembly: Preserve( (System.Linq.Enumerable), AllMembers = )]

[assembly: Preserve( (System.Linq.IQueryable), AllMembers = )] typeof true typeof true typeof true typeof true We need to tell the linker explicitly to include these assemblies, because it has no other way to know that EntityFramework Core needs them. If we don’t, the application will crash while we are migrating the database or if we try to use LINQ queries for data filtering. As a last step, open the file and in the constructor add these two lines: App.xaml.cs db = ApplicationDbContext();

db.Database.Migrate(); var new This applies all the pending migrations to the database, and if it doesn’t exist it creates the database. Creating a migration To create a migration for our app to use we cannot use any of these projects sadly, so we need to create a new project in the same solution. Let us create a console application this time that targets .NETCore so we can create a migration. The next step is to change the default namespace the project has, to match the one of the shared class library. In my example, the namespace will be changed to XamarinEFCore. , and . Then we install: Microsoft.EntityFrameworkCore Microsoft.EntityFrameworkCore.Sqlite Microsoft.EntityFrameworkCore.Design We also need to add a reference to our Entites project. Next, we will recreate the DatabaseContext folder and the ApplicationDbContext class in the MigrationCreator project. The only difference will be in the overridden method. It should look like this. protected override OnConfiguring(DbContextOptionsBuilder optionsBuilder)
{
    optionsBuilder.UseSqlite ($ );
} void "Filename={DatabaseName}" You are done! Congratulations Now you only need to create the migration and output it to the traditional shared library of the application. You do that from the command line. You navigate to the MigrationCreator folder which contains your .csproj file and execute the following command dotnet ef migrations add -o <path-to-your-shared-lib> < > MigrationName All set! Happy coding. The full source code for the application sample can be found here. Also published at: https://blog.codechem.com/using-entityframework-core-and-sqlite-with-xamarin-forms

How to Set Up a Xamarin Forms App with SQLite Local Storage

If you have ever developed an iOS app, or even started learning about it, then you must have been a bit (or a lot) confused about the process of app provisioning. I know for sure I was. So that’s why I decided to write this article explaining what in fact are these certificates and profiles, why we need them and a step by step tutorial of how to create and use them. And hopefully it will help some lost fellow out there who’s trying to find a point to this mystic process. So let’s dive into this! 👩‍💻 What are Apple certificates and provisioning profiles and why do we need them? The certificate is a digital identification you use to sign your iOS apps for development and distribution. The main part of the certificate is your public/private key pair, which is a guarantee that the apps are truly developed/distributed by you. You've probably realized by now that you can't install any app on an iOS device. It needs to be signed by Apple first and that's why we need the provisioning profiles. You can see them as a link between the device and the developer account. And if you’re wondering why is Apple requiring all of this? Well, because they are really focused towards providing secure devices. That's why their developers need to be registered, sign their apps and have them reviewed by Apple before any distribution. This is the process they came up with as a way to keep the security on a high level but still be accessible. Creating and using the certificates and provisioning profiles In order to create or use an existing certificate or provisioning profile login to your Apple developer account and you should see the homepage for your Apple developer program account. Navigate to Certificates, Identifiers & Profiles. Note: every certificate and provisioning profile have an expiration date, so it’s really important to check it before using any. Creating a certificate: In the certificates section you’ll see all of the current certificates that are made under that developer program. There are two types of certificates: and . Development Distribution The development one is used when you want to install the app using XCode (when developing an app), and the distribution one is used for distribution of the application (TestFlight/AppStore). If you want to use a current certificate then you can select it and download it, but you would also need to obtain the private key from the developer who created the certificate. 📎 Steps for creating a new certificate: select the plus icon and choose the type of certificate you need (distribution/development) and the type of service - if you have any click continue and you will be navigated to a page where you need to generate a Certificate Signing Request (CSR) file from your Mac. To create a CSR you need to follow these steps: Launch Keychain Access located in /Applications/Utilities. Choose Keychain Access > Certificate Assistant > Request a Certificate from a Certificate Authority. In the Certificate Assistant dialog, enter an email address in the User Email Address field. In the Common Name field, enter a name for the key (for example, Gita Kumar Dev Key). Leave the CA Email Address field empty. Choose “Saved to disk”, and click Continue. This example was taken from the official Apple guide You can now save the request and add it to the request field. After this the certificate is created and you can download it. . Find the downloaded certificate, double click on it and it should now appear in your Keychain The last step depends on whether the certificate is a distribution or development one. - you need to add it to your Keychain and in the Debug configuration of your project in the development environment that you use. Development certificate - you also need to add it in your project but under Release configuration (iOS bundle signing -> signing identity). Additionally, you will need to add the installation of the certificate as a task in your build pipeline (ex. Azure DevOps) before building the project. To do this, first you need to export the certificate from Keychain. Find the certificate in Keychain, right click on it and select export. It will require you to set a password for it, be sure to keep it somewhere because you’ll need it! Next, add the exported certificate with a .p12 format in the repo of the project and add it in the pipeline as well. Next, enter the password and the certificate signing identity (same as the name of the certificate in keychain). One last step is to go in the Build project task in the pipeline and add the same certificate signing identity in the Signing identity field Distribution certificate Creating a provisioning profile: The next thing we need to do if we want to develop or distribute our app is create a Provisioning profile (specific for the certificate that we just created). Each provisioning profile stores the following: certificate, an App ID (two-part string used to identify one or more apps from a single development team) and unique device identifiers (list of devices that the app can run on). Go back to Certificates, Identifiers & Profiles and under Provisioning profiles you will see all of the current profiles that are made under that Apple developer program. There are also two types of provisioning profiles: and . The development one is used for testing of the application, while the distribution one is used when we want to submit our app to AppStore/TestFlight. If you want to use a current profile then you can select it and download it. Distribution Development 📎 Steps for creating a new provisioning profile: select the plus icon and choose the type of profile that you want to create (development/distribution-app store or ad-hoc) next select an App ID - if you plan to use services such as Game Center, In-App Purchase, and Push Notifications, or want a Bundle ID unique to a single app, use an explicit App ID. If you want to create one provisioning profile for multiple apps or don't need a specific Bundle ID, select a Wildcard App ID. Wildcard App IDs use an asterisk (*) as the last digit in the Bundle ID field. in the next step choose what certificate and device/devices is this profile referred to name your profile, generate and download it after it is downloaded find it in your local storage and double click on it to install it in XCode - after successfully installing it you can add it in your project repo and in the project options under iOS Bundle Signing - Provisioning profile in your developing environment The last step here also depends on whether the certificate is a distribution or development one: - you need to select it in your project options under Debug configuration, but be careful, the profile will only appear in the selection list if you have previously selected the certificate that this profile is created for Development profile besides adding it to the project options under Release configuration, you will also need to add the installation of the profile in your build pipeline (ex. Azure DevOps) before or after installing the certificate. To do this, you just need to select the profile from the project repo in the Provisioning profile field. Additionally, under the Build project task in the pipeline you will need to enter the Provisioning profile UUID which can be found in XCode or by right click on the profile and . Distribution profile Get info As mentioned before the certificates and provisioning profiles have their expiration dates. After the expiration date has passed, you will need to create new ones in order to keep working on your app. With all of this set you're ready to go and develop or distribute your iOS app! 📲 Previously published at https://blog.codechem.com/apple-certificates-and-provisioning-profiles-a-mystery-resolved

Developing & Distributing iOS apps: Apple Certificates and Provisioning Profiles

If you’re like me, you've probably woken up on a Monday to find some of your services suddenly failing; perhaps it’s your mobile app failing to connect to your API or your images suddenly failing to load. In my case, some of our mobile apps (Android, specifically) were . After debugging for a bit I found that it was an SSL issue. reporting errors trying to retrieve data from our APIs Opening the URL in question in Chrome or making requests on iOS, however, showed no issues, and I was greeted to the normal green lock icon representing a normal certificate validation. My Issue Modern browsers and HTTP clients, like the iOS one, have better TLS trust verification strategies than older HTTP clients, like cURL and the Android built-in one. They will build a chain of trust to the root instead of stopping and failing at the first expired certificate in the chain. Chrome happily validates the certificate chain, recognizing the newer USERTrust Root Certificate. cURL, however, fails validation saying the certificate has expired. So what’s the problem and how do we debug further? Well to investigate SSL issues we can rely on good old OpenSSL. You can use OpenSSL to fetch all certificate data for a given server by running: # Hostname without protocol (no http/s)
openssl s_client -connect <hostname>: < > port The output you will receive should contain a section containing the full list of certificates in the chain: Certificate chain s: ### No is your own certificate, followed by all intermediate and root certificates s: O=AddTrust AB/OU=AddTrust External TTP Network/CN=AddTrust External CA Root
   i: O=AddTrust AB/OU=AddTrust External TTP Network/CN=AddTrust External CA Root s: ST=New Jersey/L=Jersey City/O=The USERTRUST Network/CN=USERTrust RSA Certification Authority
   i: O=AddTrust AB/OU=AddTrust External TTP Network/CN=AddTrust External CA Root s: ST=DE/L=Wilmington/O=Corporation Service Company/CN=Trusted Secure Certificate Authority i: ST=New Jersey/L=Jersey City/O=The USERTRUST Network/CN=USERTrust RSA Certification Authority 0 .0 1 /C=SE/ /C=SE/ 2 /C=US/ /C=SE/ 3 /C=US/ 5 /C=US/ Notice the in the 1st and 2nd positions in the certificate chain. These certificates are now expired and are causing errors in and the Android HTTP client. AddTrust certificate expiration cURL How do we mitigate this? The proper way to mitigate this is on the is to update the SSL configuration of the server or the reverse-proxy if using TLS-termination. server-side You can check your certificate chain; put the URL of your site into and it will spit out useful information about your certificate chain issues. You can even put in your (domain) and it will auto-remove the expired AddTrust certificates and produce a fixed .crt file you can put into your existing configuration as a drop-in replacement. https://whatsmychaincert.com hostname or you’re having issues with another platform and just want to fix your client until the server admin picks up the slack, you have a few options: If you’re not a server administrator On Linux-based client environments (Ubuntu and others), you can edit the file: ca-certificates.conf /etc/ca-certificates.conf There you will find 2 references to ; a quick fix is to just prepend! Before each line, run update-ca-certificates to apply the changes afterward. AddTrust On , a simple way to fix this for me was to edit /etc/ssl/cert.pem, find the certificate, and completely remove it. macOS AddTrust Make sure you only delete lines between and including: ### AddTrust AB
-----END CERTIFICATE----- This should fix your issues; the cURL and other HTTP clients should now be happy to resolve the new properly. USERTrust Root Certificate Hope you found this helpful and I saved you some headache, cheers! Previously published at https://blog.codechem.com/fixing-the-addtrust-external-ca-root-expiration

Fixing Root Certificate Expiration Errors in cURL

In order to submit an app to the Google Play Store, you need to sign the apk no matter what development framework you're using. In this article, we'll learn how to sign an apk using Cordova. Before starting we will assume that we have previously made all the necessary Cordova installations and added an android platform for our project. Create a keystore file First, we'll need to create the keystore file which is basically a binary file that can hold a set of keys. To create this file you'll need to go to the root of your project and run the command below: keytool -genkey -v -keystore myapp.keystore - myappalias -validity 10000 alias The tag stands for the expiration of the key. After running this command you will see the file at the root of your project. validity myapp.keystore Create a build.json file The file also needs to be created at root level and this is the place where we will store all the parameters needed for signing. The file needs to look something like this: build.json { : { : { : , : , : , : , : }, : { : , : , : , : , : }
    }
} "android" "debug" "keystore" "./myapp.keystore" "storePassword" "pass" "alias" "myappalias" "password" "pass" "keystoreType" "jks" "release" "keystore" "/myapp.keystore" "storePassword" "pass" "alias" "myappalias" "password" "pass" "keystoreType" "jks" Set version and package name Next, we need to do is open the file at root level where we can set the version, version code, and package name in the tag at the top of the file. config.xml widget < = = = = > widget android-versionCode "10" defaultlocale "en-US" id "com.example.myapp" version "1.0" ... is used only for calculations by Google Play and is not visible for users versionCode is the package name of your app id is the version of your app that will be visible on Google Play version Building the app The last thing that we want to do is build the app using the parameters that we defined in . We will achieve that by adding the flag to the build command: build.json --buildConfig cordova build android --release --buildConfig This command will build a signed apk that you can upload to Google Play. The location of the apk will be written as an output of the command, but the correct location should . be platforms/android/app/build/outputs/apk/release For more details on the installation, building, and signing you can go to the official Cordova Android Platform Guide Previously published at https://blog.codechem.com/signing-android-apk-with-apache-cordova

How to Sign Android APKs with Apache Cordova: A Brief Guide

On 20th of February 2021 at CodeChem we organized the third Virtual Open Day, which many of us will remember for a long time. It was challenging, fun and exciting. Unfortunately, due to the COVID situation we had to hold the event online. That limited us to display the chemistry that's present in our company. But we still managed to pull it off. The event was attended by more than 150 young people ready to give their best. We rolled up our sleeves, prepared a hundred virtual machines, and created a lot of discord channels. We were motivated and impressed by the number of applicants. Why Open Day is so important to us? Open Day is not just important to us, it's the most important event of the year for CodeChem. During it, we get to meet interesting people, smart students, and folks full of ambition. We get to hang out with them, teach them, but also allow them to teach us too. Everyone has a unique set of qualities. These are the goals of the event: . It's always a pleasure to meet young people who are hungry for knowledge, not afraid of new challenges and who are interested in technology. Right? Meeting smart people . This is one of our core values, but we also like to exercise it outside of the company as well. Teaching is in our DNA. It's what we love to do. Knowledge sharing through problem solving . This is how we started the company, and it has proven itself to be the best way to grow our team. Discovering like-minded people . Open Day is not only for us to find our colleagues and friends. It's an event where both sides are winning. Some candidates continue to work with us with the But those who don't, they leave smarter. Let no one leave empty-handed Boost Internship Program. A sneak peek from Open Day First, we started with a presentation about our company. It was short because there were over 150 candidates. We had to divide them into two batches because the time was limited. Then, the next 3 hours were intended for the workshop. The candidates were divided into software and embedded teams. Each team consisted of two candidates and one mentor. During which the candidates were given a task. The idea behind ​​this concept is to think together and help each other. The mentors helped them achieve that easily. The problem solving process was not easy at all. But it was a good way to show off their skills so we can test their knowledge. The candidates saw it as great challenge but they didn't give up. Kudos to them. Тhe atmosphere in our office Тwo days after the event, we asked the candidates to share their impressions through an anonymous survey. Here's a part of their answers: "It was a truly wonderful experience, something unmatched by any standard teaching or learning. I learned a lot and had a great time." "Fun! The problem I faced was interesting. Working with a mentor was a new experience, as was the whole day. I also liked the approach and structure of CodeChem." "The whole organization was great. My mentor left a very good impression on me. Now I know where I am and where I need to be career wise. It was a great experience." "It was a fun experience. Especially the pair programming part. It was interesting for me to adapt to someone's tempo and mindset. Also finding ways to explain my point of view, and come to a common solution." "Open Day is a great way to relieve the stress of a traditional interview with definitions." "For the first time I participated in this kind of Open Day. But the experience is only with positive feedback. I've had An interesting connection with the employees. I saw the way you should communicate with them. That was a new experience for me. That taught me how you should function in a team or in a given task that you need to perform. Plus, I realized what level my knowledge is and what I need to correct or upgrade. Thank you for the opportunity to attend such an event." "Before the event, I had no experience when it comes to direct contact with an IT company. I was skeptical about working in a team but I was surprised by the good organisation of the CodeChem team. I was in the Software group. The application was quite interesting. I could see where I made mistakes and where I needed help. You inspired me to make Bejeweled in React! Very pleased with the event!" "Virtual Open Day 2021 was a very unique experience for me. The cooperation with the mentor was great. I learned from the open discussions to look at the problem from another perspective. We discussed different implementations and what is a better solution. The relaxed atmosphere made the event fun and interesting." "Great! I like that we worked in a team + mentor. We had great communication and a good problem to solve. I wanted it to last a bit more." "The idea for the event itself was very good and everything was well organized. The mentor was attentive, patient and very dedicated. All in all, it was a nice experience to work in a team and share knowledge and ideas. Thanks to the whole team for the opportunity to attend such an event." "Interesting and unusual. This is my first time attending this type of event. The problem was interesting. Everyone from CodeChem looked very nice and approachable." "The environment was very sociable and my time passed very quickly. The event was engaging and, above all, interesting." "The online event was great. It was well organized and the mentors were available at all times. There was a good atmosphere. The task (at least the one for embedded group) was interesting and challenging." "CodeChem left a great first impression on me. The employees had a friendly relationship with each other. That made me want to be part of such an environment. I hope the mentors are always in the same mood for helping and explaining as they were on the Open Day." "At the beginning it looked like an ordinary IT company. But, as I watched the synergy between the team I realized that this is a very interesting approach. I hope one day I could be part of that team." It was our first experience with a Virtual Open Day. We didn't know what to expect and what outcome we would have. But now, we can proudly say that the Virtual Open Day 2021 brought us a lesson, new friends, and tons of experience. We're looking forward to organizing the next one, but hopefully in real life, like we've always used to do. Previously published at https://blog.codechem.com/we-mentored-over-150-students

Virtual Open Day 2021: Mentoring Over 150 Brilliant Students Virtually

What is Hyperparameter Optimization? In machine learning, hyperparameter optimization or tuning is the problem of choosing a set of optimal hyperparameters for a learning algorithm. A hyperparameter is a parameter whose value is used to control the learning process. By contrast, the values of other parameters (typically node weights) are learned. Strategies Several strategies can be used for performing optimization. The most simple one is manual tuning. One such example is using the [Elbow Method](https://en.wikipedia.org/wiki/Elbow_method_(clustering) for determining the number of clusters in . On the other hand complex models, have dozens of hyperparameters, and combined with the fact that some of them are continuous, the size of the search space explodes, so the manual effort. To tackle this issue, several other "smarter" approaches exist. Some of them are: k -nearest neighbors algorithm Grid search Random search Bayesian optimization Gradient-based optimization Evolutionary optimization Population-based To get more familiar with how these approaches work and what are the differences between them, I encourage you to read these articles , , or . 7 Hyperparameter Optimization Techniques Every Data Scientist Should Know How To Make Deep Learning Models That Don’t Suck Hyperparameter Optimization Approaches Tooling Several frameworks provide implementations of the approaches mentioned above. In this tutorial, we are going to explore , . Weights & Biases - Sweeps (WANDB for short) Setup For this tutorial, we are going to build a classifier for the dataset. We will use from sklearn to predict the presence of heart disease. Heart Disease UCI RandomForestClassifier This tutorial does not focus on data pre-processing, so we'll dive straight into splitting the data into train and test data, and train the model once with default values for hyperparameters. import pandas as pd
from sklearn.model_selection import train_test_split

df = pd.read_csv('data\heart.csv')

X = df.drop(['target'], axis=1)
y = df['target']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2) Train the model rfc = RandomForestClassifier()
rfc.fit(X_train, y_train) and that's it, now we have successfully trained a random forest classifier with default values for its hyperparameters. This classifier has the following hyperparameters: bootstrap max_depth max_features min_samples_leaf min_samples_split n_estimators Now let's get into setting up the optimization: Optimization Step 1: Define the Training Script For more details, see the . docs This script should serve as the main entry point for optimization. It performs one training and evaluation of the model with values for the hyperparameters injected from outside (through ). wandb.config It gets the configuration from the outside and performs training and evaluation of the model with fixed values for all of the hyperparameters. The name could be arbitrary, and for this example is . train.py import wandb
import pandas as pd
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score
from sklearn.model_selection import train_test_split

WANDB_PROJECT_NAME = "hyperparameter-optimization"

with wandb.init(project=WANDB_PROJECT_NAME):
    df = pd.read_csv('data\heart.csv')
    X = df.drop(['target'], axis=1)
    y = df['target']

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

    config = wandb.config
    rfc = RandomForestClassifier(
        bootstrap=config.bootstrap,
        max_depth = config.max_depth,
        max_features = config.max_features,
        min_samples_leaf = config.min_samples_leaf,
        min_samples_split = config.min_samples_split,
        n_estimators = config.n_estimators,
    )

    rfc.fit(X_train, y_train)
    y_pred = rfc.predict(X_test)

    wandb.log({'accuracy': accuracy_score(y_test, y_pred)}) Step 2: Define the Optimization Strategy and Configuration To determine the optimization strategy, i.e. which approach to be used for optimization, what values or ranges should be tried for every hyper-parameter, and what objective to be optimized, a configuration file needs to be defined. sweep.yml This file contains some additional configuration regarding the python path, training script path (from Step 1), to get more familiar see the . docs For this example the file is. sweep.yml program: train.py
method: bayes
project: hyperparameter-optimization
command:
- ${env} 
- ~/envs/hyperopt/bin/python
- ${program}
- ${args}

metric:
  name: accuracy
  goal: maximize
parameters:
  bootstrap:
    values: [True, False]
  max_depth:
    values: [2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, None]
  max_features:
    values: ['auto', 'sqrt']
  min_samples_leaf:
    values: [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
  min_samples_split:
    values: [2, 3, 4, 5, 6, 7, 8, 9, 10, 11]
  n_estimators:
    values: [10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 500] Interpretation is that the training script is located at , Bayesian optimization is going to be used and for bootstrap, the values and `False is going to be tried, for max_depth [2, 3, 4, .... ] and similarly for all of the hyperparameters. The objective is maximizing accuracy. So in other words, . train.py True we want to assign values to the hyperparameters such that the accuracy is maximized Step 3: Initializing and Running the optimization To start optimization, open a shell (with your favorite terminal emulator). Activate the python virtual environment (for UNIX , for other see the official Python .) source ~/envs/hyperopt/bin/python guide Initialize the sweep wandb sweep .\sweep.yml Expected Output: Run the sweep wandb agent aleksandar1932/hyperparameter-optimization/r3s5xf4d Expected Output: And that's it, now the running sweep can be observed at WANDB. Step 4: Monitoring Go to the sweep URL, from your shell output. For this example, the output is available , and below. here As the model is trained for different combinations for the hyperparameters, the results are updated in real-time. We can wait for the given approach to find the combination of values that when used for training, provide the best model, or stop the optimization either by terminating the running shell or through WANDB. Conclusion This sweep performed 79 runs, and the best model scored accuracy on a randomly sampled test set for that particular run. train_test_split is done for every run) 0.9016 (In train.py So, it can be concluded that the best RandomForestClassifier should be instantiated with the following hyperparameters. model = RandomForestClassifier(
        bootstrap=False,
        max_depth = 4,
        max_features = 'sqrt',
        min_samples_leaf = 3,
        min_samples_split = 4,
        n_estimators = 200,
    ) The code from this tutorial is available on . GitHub First published here

Using Weights and Biases to Perform Hyperparameter Optimization

In our latest blog post, we've discussed what a is. It's the easiest tool you can build in order to test your idea and learn through collecting feedback from your early adopters. And also, how to build it. Minimal Viable Product It's a fact that in this dynamic and turbulent world, full of brilliant ideas ready to be born into great startups, every entrepreneur is facing the same challenge - a blank page! That is one of the main reasons why many startups are failing. The blank page has endless possibilities and no guidelines. And the absence of form scares off potential entrepreneurs before they start. The absence of form exists because of the lack of feedback, so entrepreneurs don't know whether they are making the right decisions. Successful entrepreneurs see the blank page differently. They will shape a blank page into many ideas, then a plan, and at the end a product. They will transform the page into a process that will allow for that evolution to occur. Successful entrepreneurs don’t rely on the intuition of what users want, instead, they learn it through a process of hypothesis and experimentation. They discover the best decisions for their business by building an MVP. MVPs can come in a thousand shapes and forms. So, in this article, we will focus on the most popular ones. We will give you a simple explanation and examples from successful startups. We're sure that it will help you decide which type of solution is most suitable! Content-related MVP This group of MVPs is based on content. It will allow you to present the idea behind your product to your potential users or investors without building a prototype. And now you probably have a question: ‘Isn't MVP first and foremost a product?’ Yes! But, it’s possible to catch potential leads and measure purchase intent without building it. A landing page is the most used form of MVP because is easy to make. A simple website, a short description of your product, and a CTA (call to action) button can be a very reliable test of people’s interest. For a quick result – you can run an AdWords campaign on your main keywords and see people's reactions. Main tips: ✔︎ Make it look aesthetically appealing, so you can grab people’s attention! ✔︎ Try to explain all the features from the user’s perspective. ✔︎ Ask people to leave their e-mails. Once the MVP is live, you’ll be able to inform early adopters about your product. Example: Buffer.com One of the most popular platforms for scheduling social media posts was tested through various landing pages before the first line of its code was written. Joel Gascoigne, the founder, build a simple website describing Buffer’s functionalities and a page with “Plans & Pricing”. Everyone who clicked on could be a potential subscriber. Most importantly, the product wasn’t ready yet. The page only contained a piece of information that it will be launched soon and a possibility of signing up for a newsletter. Very easy and effective! 🤩 Pros Can be built cheaply and quickly. Can easily be matched with online ads. Can be tested and optimized easily. 😩 Cons The conversion rate is low (about 1-3% of people sign up). Hard to adapt all vital information into one page. A landing page that's not aesthetic can hurt your brand. This type of MVP is another popular content-related MVP. One minute-long video is more than enough to explain your product. An interesting and creative video catches people's attention more easily than a simple description. Also, can go viral very quickly. Example: dropbox.com https://www.youtube.com/watch?v=w4eTR7tci6A DropBox (a platform for sharing files on different devices) proved the effectiveness of their MVP in the form of a video. Because of the complexity of their innovative product, they test it before investing resources for development. DropBox created a short animated video for their MVP. The subscriptions sky-rocketed overnight! 🤩 Pros Explains your product in a simple way. People are interested in watching a short video. Good for branding. 😩 Cons You may have to spend a lot of time making an explainer video and getting the right message. Functionality-related MVP This type of MVP is based on functionality. Can be used when content-related MVP wouldn’t give you enough feedback from your early adopters. Wizard of Oz MVP is where you essentially put up a front that gives your early adopters the impression that you have a product. Also, users have the opportunity to experience it. It’s a very effective way of checking if you have a wanted product before you actually build it. Example: zappos.com Zappos (an online shoe retailer) is the most famous example of Wizard of Oz MVP. Nick, the founder tested the idea before creating the product. He went to local stores, photographed shoes, and sell them online. All that he needed was a very simple website and a small budget for ads to get traffic. From the customer’s side, it looked like a fully functional website. 🤩 Pros You can set up this MVP cheaply and quickly. You can use social media to generate interest. 😩 Cons You have to spend a lot of time and effort to manually perform the service. Customers can feel cheated if they come to know that there is no real product. Single Feature MVP is when you develop your actual product and sell it to your potential users. Even though the term means developing 'one single’ feature for the MVP, it doesn’t necessarily mean that. Single Feature MVPs are those that focus on solving the core problem of the user. And to do so, you may need more features in your minimum viable product, but they must be prioritized. Keep ‘must-have’ features and eliminate ‘nice to have’. While defining and prioritizing the core features to build, consider the following things: Unique Selling Point that offers users value and stand among the competition. Leave the other features for later development. Example: spotify.com Spotify today is the global leader in the music industry. But how it reached the top of success is an interesting story. In 2008 Daniel Ek and Martin Lorentzon identified the opportunity that the music industry had nothing to help the music lovers stream songs. In 2006 after just 4 months of development, they launched their first prototype. 🤩 Pros Focused on solving one specific problem for specific users. You can get your product to market fast with low costs. You can expand it later without much difficulty. 😩 Cons You have to invest in development. You may have a problem figuring out which feature you should focus on. Before choosing the right type of MVP for your startup, you should define your goal. Understand your target, estimate the resources, choose the most suitable type MVP for your startup, and get ready to beat the market. As a software development company, we are always here to help you make the right decision and turn your ideas into a successful startup! ★ NEXT STEP ⇢ In this post, we went through what are the most popular types of MVP. In the following post, we will explain how to Stay tuned. define and prioritize the features for your MVP. First published here

How to Choose the Right Type of MVP for Your Startup Idea

Back in October 2020, I decided to start a pet project with a goal to learn new technology. I noticed people on Twitter were saying how much they like the Developer Experience of Blitz JS so I decided to use it in this pet project. But then I thought, this pet project could be something useful to our employees. So, after a little research, I decided to build an Absence Tracker, so we can ditch the Telegram group where we reported our absences. The Inception I decided to use along with , because it's rich with components and it allows me to quickly build the UI. And it looks 😍. I did a little planning and Database Design in and started to layout the pages. Chakra UI Blitz JS diagrams.net One month in I decided to stream the whole development process on ! That was one of my best decisions. I've met a lot of cool folks on my streams and they gave me some interesting ideas regarding the implementation. All of my streaming archives can be found on my . Twitch YouTube channel The Challenges One of the challenges was keeping up with the changes of Blitz JS. When I started the project Blitz was still in alpha. And as we all know, alpha software tends to change and it's totally normal. Blitz JS is maturing in a great way, but keeping track of the changes did require a lot of release notes reading, and checking out past PRs. Another challenge was using Chakra UI with an experimental version of React (default by Blitz JS). I couldn't find a good workaround, so I decided to use the stable version of React. The drawback of that was disabling the Concurrent mode of React, even though I really liked that feature. The Benefits This tech stack grew to be my favourite. It really allows you to build your app in record time! I really like the DX of Blitz JS and Chakra UI and their intuitive APIs. I'll definitely use these technologies for my future projects too. As I mentioned, streaming on Twitch was one of my best decision. I've gotten a lot of followers and met a lot of cool folks that hung around my streams. I've also gotten the attention of Chakra UI! They liked my streams and invited me to join the 🤩 core team The Finale I decided to bring my colleague as a guest to my live stream where we Dockerized the app and deployed it on our company servers. There was a ready-made Docker image for Blitz JS, but we decided to build our own. It wasn't complicated at all. After we set up everything in a Docker Compose network, the deployment was a single line command. Ilija Boshkov Next on our list is to onboard all of the employees, synchronise the old data with the new platform, and officially start using the new platform. It started off as a pet project, a "placeholder" for me to learn new technology, but it grew into a cool usable app that will make our lives easier. It's been a wild ride, but I'd do it all over again without thinking.

Stories, News & Mentions about CODECHEM

Categories

Trending Topics

Stories, News & Mentions about CODECHEM

Codechem’s Stories on HackerNoon

Categories

Trending Topics