The Pro model has electromagnetic dimming, which allows it to change the transparency of the glass - from fully transparent to fully opaque.

Xreal Air 2 and Air 2 Pro glasses go global with better displays and lower weight

Get the most recent info and news about AirFit on HackerNoon, where 10k+ technologists publish stories for 4M+ monthly readers.

AirFit

44,000 startups. 4,553 cities. “4 billion votes” It was such a blessing to get to know so many of your ideas and innovations from every corner of the world in 2021. The startup community amazed all of us with your hard work and persistence despite all of the challenges that we faced globally throughout last year. Not gonna lie, the competition got our hearts racing like Eurovision. We won’t let you wait for much longer, below are long-awaited results! ✨✨✨ To see who is a winner in each respective city, visit and pick a city. The name at the top of the list is the winner. (Compromised awards are under review.) startups.hackernoon.com The will announce the winners per region in their respective cities. Startups Winners Announcement Series 🔥 Central Europe Central Europe has always caught our attention with the talented pool of people in tech community. But were they able to get through 2021 despite the pandemic? Let’s cheers to the most outstanding amongst 4,500+ startups who have been working hard this year: Basel, Switzerland An overwhelming vote for ~ in their mission of helping machinery companies drive more service revenues from their installed base MachiQ Software Berlin, Germany We have with a whooping 538 votes! Let’s support Plan A with its mission to help companies to measure, monitor and reduce their environmental footprint. Plan A Düsseldorf, Germany It doesn’t get any easier with Kicking off in helping people organize online schedules more easily. EasyWeek! Geneva, Switzerland Manage code quality? We have to give props to for their startup! SonarSource Hamburg, Germany Taking over Hamburg is , an acoustic device to reduce noises in vehicle cabins. No more annoying sounds on the road 🚗 ReCalm Köln, Germany Small business? Hoping to manage your marketing campaign? is winning the vote! rankingCoach Lausanne, Switzerland is a Swiss-based company focused on cryptocurrency management solutions built for financial institutions. Metaco Leipzig, Germany empowers companies to operate and utilize drones for the most effective industrial data applications. FlyNex Mainz, Germany is growing sustainable digital mobility startups! Next Mobility Labs Munich, Germany wins for top-line, high-security international Identity Management! ORY Stuttgart, Germany Remote desktop application? Ask anywhere. AnyDesk Vienna, Austria , an IT company enabling privacy-preserving big data applications in the digital age. Mostly AI Zug, Switzerland Way to go 1762 votes for the first provably live pure Proof-of-Stake (PoS) CBC-Casper blockchain designed for real-world applications. CasperLabs! Nyon Calling for robust data solutions for impact-driven organizations! Havuta LLC Zürich is revolutionizing small business lending to give easy access to capital to those who need it most: SMEs ✨ Teylor Heidelberg, Germany is an innovative HR startup providing a solution for fast and efficient recruiting of employees with simple job profiles. Talk'n'Job Stuttgart, Germany Congratulations for winning this title! AnyDesk Moving on to our buddies in... 🔥 Eastern Europe Eastern Europe is home to many familiar big names in our start-up world. Trace history for Miro, Telegram, and Grammarly and you know where they are from. Who is the next generation for Eastern European startups? Budapest, Hungary is a platform for mobile app development, enabling tens of thousands of developers to automate integration, testing, and deployment. Bitrise I__ __ stanbul, Turkey Leading here is - a brand new software intelligence platform- helps organizations deliver faster, high-quality products by connecting their DevOps tools. Oobeya Kraków, Poland is a human cloud platform that connects global enterprises with European software houses, offering tech talent as a service. Talent Alpha Kyiv, Ukraine 281 votes for , an AI powered app where people can swap faces in the videos, GIFs and images in just a few seconds. Reface Sofia, Bulgaria 387 votes for __ __378, leading Blockchain and DLT solutions company that specializes in helping enterprises and blockchain startups design, build and scale innovative solutions using Blockchain & DLT! LimeChain Warsaw, Poland is a next-generation blockchain document management software (DMS). DoxyChain Yerevan, Armenia Whoever clicked to vote for was sure a vote, because it ended up winning as a full-service software provider delivering Artificial Intelligence & Machine Learning solutions for businesses. SmartClick smart Almaty Just like the name suggested, provides one pass to dozens of gyms and studios. 1FIT Debrecen got all the coders ranked figured out! CodersRank Kharkiv is complete solution digital assets management! Pics.io Minsk, Belarus with 243 votes is the first blockchain-powered sports media platform. Scorum Moscow, Russian Federation 1 e-grocery company in Russia with 26 votes and #1 on our HackerNoon board for Minsk. SberMarket Saint Petersburg, Russian Federation leads this area as a personal finance application that rewards responsible spending and savings. CoinKeeper Kazan, Russian Federation Smelling a sweet victory? 3d App designers for flower bouquets is in the house! Flowito Western Europe 🔥 Brexit? Pandemic? It seems like startups in Western Europe still keep their spirits regardless. Startups in Western Europe that stood out in votes this year go to: Antwerp is on a streak! As a bug bounty platform, it ensures online security through a crowd-sourced platform for computer experts. intigriti Belfast A record-breaking 10,003 votes? is an independent coffee shop. Come grab a coffee! Bobs & Berts Birmingham What’s happening in Birmingham? We have , an innovative UK-based advanced lithium battery developer aiming to accelerate the global shift to cleaner. Aceleron Bordeaux 30 is an open-source web framework for building modern apps and websites with Vue.js. NuxtJS Boulogne-billancourt is a software company that provides an in-store analytics platform to optimize operations. DiGEiZ Brighton In for some jam? offers musical education services. BIMM Bristol Another tasty victory for Western Europe is enabling the strongest digital encryption on the internet. KETS Quantum Security Brussels A fashionable win for , a social fashion network. You scan barcodes in fashion shops and see every influencer wearing those outfits! Share your wishlist with your friends and earn cashback. SCANDALOOK Cardiff is a technology company developing end-to-end digital connectivity for the point-of-care diagnostics market. Bond Digital Health London A whooping number of 1085 vote__ __ offers hardware based cyber-security solution for authentication and encryption in a wired local network. Crypto Quantique Edinburgh Hopping up on the crypto movement, just launched last year but they are already rocking it at making crypto accessible to everyone. Zumo Ghent 5379 votes for ! A Belgian InsurTech startup that helps insurance providers all over the world to better connect with their digital customers. WeGroup Glasgow specializes in single-cell analysis transforming the life sciences. Biomage Paris Show some love in the city of love for , a sports platform that is connected to a mobile sports coaching application. AirFit Aberystwyth enables artificial intelligence, data science & digital innovation solutions for the agriculture, biotech & life sciences industries. Agxio Norwich Congratulations for the amazing win of 1124 votes! WATR Charleroi Congratulations for your amazing win! Hirering Gibraltar Give a round of applause to Fortknoxter! Waltham Abbey is ruling the area with its amazing 72 votes! RPAIR Antwerpen Rocking up Antwerpen area is the rising startup ! Willow Egham What’s happening in Egham? Proudly announcing winning the Noonies this year! ATA.ONE Leicester Leicester is calling the name of as the winner for the region! DeadHappy Macclesfield Who else but ? Congratulations on dominating the award! AeroCloud Northern Europe 🔥 Want to listen to some music? Spotify. Need some games to kill time? Candy Crush. What about a quick call to someone you haven’t seen in a long time? Skype. These are all startups that originated from Northern Europe! Noonies, let’s take a look at the upcoming generation of startups from Northern Europe that hopefully will see an even bigger and brighter future. Gothenburg, Sweden Shipping was never made for small shippers, so took on the mission to built an end-to-end logistics platform to empower small businesses to spend less time and money on shipping and more time on growing their businesses. Sendify The Netherlands provides a virtual platform for many kinds of conferences and events. Let's Get Digital Helsinki, Finland A curated collection of public basketball courts in the Helsinki Metropolitan Area called . Everyone is welcome to submit to their local court. Koripallopaikat Lund, Sweden develops an antibiotic-free, safe and effective treatment of vaginal fungal infections. Gedea Biotech Luxembourg, Luxembourg 🚀 is a job-matching platform that connects developers and tech experts with CTOs and team leads from the best tech companies across Europe. 👾 nexten.io Malmö, Sweden Feeling studious, Noonies? __ __e is a tool for Google Classroom, that enables formative matrix assessment towards course learning objectives directly. StudyBe Oslo, Norway is the platform supports a unified API across all connected device categories, removing significant friction for app developers. Domos Labs Reykjavík, Iceland __ __s is a mobile-first technologically advanced neo-bank. Indo service #S__ __ tockholm, Sweden is the biggest blockchain start-up in Sweden. Double winner of the most prestigious nomination in Nordics 'NyTeknik 33-listan', winner of the Deloitte Fast 50, and now Noonies Award! ChromaWay AB T__ __ allinn, Estonia Passing 1000 votes is a fully digital and decentralized ecosystem for all actors in supply chain. OBORTECH Vilnius, Lithuania is an easy-to-use money app for cards and accounts. Via Genome, users can open a business or a merchant account or a personal one with dedicated IBANs, access transfers, exchange currency, process payments, and manage the funds. Genome Aalborg Digura A legaltech offering tenants fast, digital and reliable legal assistance Aarhus Partisia Market Design provides secure, simple, efficient solutions for auctions and exchanges realizing the potential of the market. Helsingborg is a hi-tech company developing cardiac occlusion devices and other vascular implants. Occlutech Tallinna is an AI-powered system for marketing automation in social networks. POSTOPLAN 🔥 Southern Europe Southern Europe’s economy went through a hard time this year due to a huge reduction in tourism. But our startups are of course always headstrong, from Spain to Italy and all the way through Greece. Who are the heroes rising amidst changes of Southern Europe? Skopje is a versatile software development company that solves the business challenges of SMBs, big brands, and brave hearts with awesome ideas. CodeChem Sliema Sounds like just a relaxing exhale. is a holistic health and wellbeing app built to allow users work on the development of their physical, mental and spiritual health Few Tajonar is a company of experienced business & industry entrepreneurs offering solutions to industries based on cutting-edge technologies. das-Nano Tarragona is a technology-based company focusing on the production of integrated microsystems. IMICROQ Terni is an ICT-based integrated solution that is able to modulate, respond, assist and learn to control most home activities. Mind Tirana is a video content and e-commerce platform for the Balkans built on top of an Albanian language specialized search engine. Gjirafa Torino develops and commercializes algorithms based on Artificial Intelligence to increase the autonomy of small satellite missions. AIKO T__ __ orres Buy online groceries like you would buy offline with personalized picking as fast as in 4 hours with Vedras 360hyper! Trento is a music-based platform built around streaming platforms that allow users to discover new emerging artists and make new friends. Brots 2021, what a YEAR. Congratulations to all of the startups that made it to the top winning list for Europe! Stay tuned for Startup awards in 2022 and let’s cheers to a new year of more innovative ideas, bold innovations, and robust creations from you, Europe 🌷 More Information: If you’re a winner, please answer ! If you’re a reader, participant, just a curious individual, consider answering one of our interviews . this interview here Read the FAQs about or . Startups here here To see all of the winners you may visit , pick any city and view the name that appears at the top. this site If you’re interested in partnering with us, we invite you to look through some of these or reach out to our business department. unique sponsorship options Perhaps creating your very own tickles your fancy or maybe you have a whole new idea for how we can partner! writing contest Thank you to all of our sponsors! We truly appreciate your contribution to the tech world and thus the future! The Startups Awards are sponsored by: , , , , , , , , , , Blockster Brex FreeCodeCamp LaunchDarkly Coil Udacity Udemy Vercel Asia Pacific Transgender Network Bybit Everscale Network This post was written by An Le and edited by Natasha Nel. If you have any questions or concerns or would like to report a discrepancy please email ellen@hackernoon.com.

Meet the Winners of Startups in Europe

Computer vision remains an extremely compelling application of artificial intelligence. Whether it’s recognizing elements on a battlefield or predicting crop yields, computer vision is arguably one of the most commercially valuable (and societally important) realms of AI. However, a rate-limiter to the adoption of the best computer vision capabilities is oftentimes the complexities associated with constructing a dataset and designing a simple end-to-end system that will perform your computer vision task on a new image. In this blog post, we will take a step-by-step look at how to tackle these issues with best-in-class tools like CVAT and MinIO Bucket Notifications. By the end of this post, you will be able to train an object detection model on a custom dataset and use it to make predictions whenever a new image appears. The Premise Let’s say we want to be able to recognize types of aircraft present in satellite imagery. Let’s also assume we are starting from scratch: no pre-built datasets, no pre-trained models. Here are two sample aircraft that we want to detect and recognize in our satellite imagery: The steps outlined in this post can be generalized to just about any domain. Instead of detecting aircraft types, we could be classifying land use or performing regression to predict crop yields. Beyond traditional images, we could also be training and performing inference on other types of multi-dimensional data like LiDAR point clouds or 3D Seismic Images; it becomes just a question of how the training data looks (and potentially a different deep learning model instead of YOLO). If you have more questions about what this would look like for a specific use case, feel free to make an issue on the GitHub ! repo Step 1: Acquiring and Managing Training Samples For this project, due in large part to not owning an on-demand imaging satellite, I visited airfields on Google Earth and took multiple screenshots of areas that had some of these planes visible. Assembling this set of images took quite some time, so I stored them all in a bucket on my MinIO server titled “object-detection.” In a production setting, the benefits of storing your collected samples on MinIO become even more prescient. Active-active replication, the highest levels of encryption, and super fast GET/PUTs (to name a few) mean that your diligently-collected samples will be highly available, safe and secure. Step 2: Creating the Dataset In order to train an object detection model for your use case, a labeled (or ‘annotated’) dataset is needed. A great tool for this is CVAT by OpenCV. A cool feature is that CVAT provides a utility to connect your MinIO bucket as a “cloud storage” in order to feed your bucket’s images directly to the dataset annotation tool. To do this, make sure that your MinIO Server’s host is accessible to the CVAT server, especially if you’re running MinIO Server on-premise or locally on your laptop. Also, as a note, there are two ways to use CVAT: (1) using their provided web app at or (2) running it locally. In either case, once you have CVAT open, click on “Cloud Storages” in the menu bar. From there, you can fill out a form to attach your (S3-Compatible) MinIO bucket: app.cvat.ai Let’s now create our new labeling task under “Tasks”: You should be prompted with a form to fill out: When creating the task, it’s important to define the class labels correctly (I defined two Rectangle labels titled “SU30” and “TU95,” corresponding to the two planes I wanted to detect): Now the remaining step is to attach our previously added MinIO bucket as a data source. Under “Select Files,” click “Cloud Storage” and fill in the name you provided for that source earlier. I used the name “minio-cv-bucket” above. The uploading process will take a few minutes. Once it’s completed, you should be able to see your annotation job available under “Jobs.” Now, by clicking on the job, you can start annotating each of your images. Warning: this can be a disproportionately time-consuming process. Generally, in a production environment with large annotation needs, it may be best to offload this task to a dedicated internal team or a third-party data labeling company. Once you’re done annotating, export the dataset in the YOLO format. Step 3: Organizing the Training Data Your exported dataset will be in the form of a zip file. Once you unzip it, the YOLO-formatted annotation text files will be within an enclosed folder. Feel free to take a look at them. In the YOLO format, each image’s annotations are in a text file where each line contains two corners of a bounding box and the class. The class number corresponds to the order in which you defined the labels when creating the task. So, in this example, 0 would correspond to Su-30, and 1 would correspond to Tu-95. At this point, create a new working directory (or enter one that you have already created). Within this directory, create a subdirectory called 'dataset.' Within ‘dataset’, create directories such that your working directory looks like this: my_cv_project (WORKING DIRECTORY)
|---- dataset
	|----images
		|----train
		|----val
		|----test
	|----annotations

		|----train
		|----val
		|----test You will now have to populate the train, val, and test subdirectories for both images and their corresponding annotations (the text files). It’s up to you how you want to retrieve and split up your samples. A good practice is to break up your total amount of training samples into 80% training, 10% validation, and 10% test. Make sure to randomly shuffle your images before partitioning them. Personally, I used MinIO Client’s in the command line to quickly retrieve all the images from my ‘object-detection’ bucket. Alternatively, if you have all your sample images in one place on your local computer already, you can directly work with that. Once I had all my samples in one place, I used a Python script to shuffle, split, and move my images and annotations to the train, val, and test directories. . If you have any questions about how to use it, feel free to make an issue in the ! mc cp Here is the script provided for convenience repo Ultimately, make sure that for each image that you have placed in either images/train, images/val, or images/test, the matching annotation .txt file is also in the corresponding subdirectory within the annotations/ directory. For example: my_cv_project (WORKING DIRECTORY)
|---- dataset
	|----images
		|----train
			- 5.png
			- 3.png
			- 2.png
		|----val
			- 4.png
		|----test
			- 1.png
	|----annotations
		|----train
			- 5.txt
			- 3.txt
			- 2.txt
		|----val
			- 4.txt
		|----test
			- 1.txt Now, our data is in place. It’s time to take a look at our object detection model and start training. Step 4: The Object Detection Model The current gold standard (in terms of performance and ease of use) for object recognition is the YOLO (You Only Look Once) class of models. At the time of writing, YOLOv8 is the latest version and is maintained as open-source by Ultralytics. YOLOv8 provides a simple API that we can leverage to train the model on our newly created annotations (and eventually run inference as well). Let’s download YOLOv8: $ pip install ultralytics We can now use the YOLOv8 CLI tool or Python SDK to train, validate, and predict. Refer to the YOLOv8 docs for further information. Step 5: Training In your working directory, define a YAML file that specifies the locations of the dataset and the details about the classes. Notice how the paths are the same as the ones I created earlier in the working directory. I called my file ‘ .’ Also, note the two aircraft class labels have to be defined in the same order as they were in CVAT. objdetect.yaml train: ./dataset/images/train/ 
val:  ./dataset/images/val/
test: ./dataset/images/test/

# number of classes
nc: 2

# class names
names: ["SU-30","TU-95"] Start training the YOLOv8 model on our dataset with the following command (using the YOLO CLI tool). Refer to the YOLO docs to learn more about all the different options you can configure for training. Here, I am initiating training for 100 epochs and setting an image size of 640 pixels (all our training images will be scaled accordingly during training): $ yolo task=detect \
  mode=train \
  model=yolov8s.pt \
  data=objdetect.yaml \
  epochs=100 \
  imgsz=640 Training will take a while, especially if you’re working on a laptop (like I am), so now is a good time to take a break (or read ahead 😀)! At the end of the training loop, your trained model, along with other interesting graphs and charts, will be stored in an auto-generated directory called ‘runs.’ The terminal output (like below) will indicate the specific location of the results of the latest run. Every time you train a model, a similar directory will be generated within ‘runs/detect/’. Results saved to runs/detect/train Note: runs/detect/train/weights/ will contain the PT files with the exact trained weights. Remember this location for later. Step 5B: Model Validation and Test You can run validation with the following command: $ yolo task=detect \                                       
  mode=val \
  model=path/to/best.pt \
  data=objdetect.yaml The results will be automatically stored in a folder in your working directory with a path of the form ‘runs/detect/val.’ To perform inference on the test set, you can use the following command: $ yolo task=detect \                                                    
  mode=predict \
  model=path/to/best.pt \
  conf=0.5 \
  source=dataset/images/test The results will be stored in ‘runs/detect/predict.’ Here are some prediction results on the test set: Step 6: New Image Inference using MinIO Bucket Notifications Now that we have a trained model that can recognize some aircraft types present in a satellite image, ? how can we make use of it for new images in a simple way is a perfect tool for this. We can build a system that can automatically perform object detection inference upon a new image dropped in our bucket with the help of a webhook. MinIO Bucket Notifications At a high level, we have 3 steps. First, we need to define an endpoint that can serve as a webhook to perform object detection on a new image with our trained model. Second, we need to configure some environment variables for our MinIO Server deployment that instruct it to hit our webhook endpoint upon some event happening. Third, we need to configure what types of bucket events (i.e. PUT) we want to act on. Let's walk through it step-by-step. Here’s the code for a simple Flask-based server ( ) that runs inference on a new image added to the MinIO bucket: detection_server.py """
This is a simple Flask inference server implementation that serves as a webhook for the event of a new image being added to a MinIO bucket. Object detection using YOLO will be performed on that image and the resulting predictions will be returned.
"""
from flask import Flask, request, abort, make_response
from ultralytics import YOLO
import tempfile
from minio import Minio

# Make sure the following are populated with your MinIO details
# (Best practice is to use environment variables!)
MINIO_ENDPOINT = ''
MINIO_ACCESS_KEY = ''
MINIO_SECRET_KEY = ''

model = YOLO('/PATH/TO/best.pt')  # load a custom model (path to trained weights)

client = Minio(
    MINIO_ENDPOINT,
    access_key=MINIO_ACCESS_KEY,
    secret_key=MINIO_SECRET_KEY,
)

app = Flask(__name__)

@app.route('/', methods=['POST'])
async def inference_bucket_webhook():
    """
    This endpoint will be called when a new object is placed in your inference bucket
    """
    if request.method == 'POST':
        # Get the request event from the 'POST' call
        event = request.json
        bucket = event['Records'][0]['s3']['bucket']['name']
        obj_name = event['Records'][0]['s3']['object']['key']
        with tempfile.TemporaryDirectory() as temp_dir:
            temp_file_name = temp_dir+'/'+obj_name
            client.fget_object(bucket, obj_name, temp_file_name)
            # See https://docs.ultralytics.com/modes/predict/ for more information about YOLO inference options
            results = model.predict(source=temp_file_name, conf=0.5, stream=False)
            # A list of bounding boxes (if any) is returned.
            # Each bounding box is in the format [x1, y1, x2, y2, probability, class].
            result = {"results": results[0].boxes.data.tolist()}
            print(result)
            resp = make_response(result, 200)
            return resp
    else:
        abort(400)

if __name__ == '__main__':
    app.run() Let’s start the inference server: $ python detection_server.py
 * Serving Flask app 'detection_server'
 * Debug mode: off
 * Running on http://127.0.0.1:5000
Press CTRL+C to quit Make note of the hostname and port that the Flask application is running on. Next, let’s start working on configuring the webhooks on the MinIO side. First, set the following environment variables. Replace <YOURFUNCTIONNAME> with a function name of your choosing. For simplicity, I went with Also, make sure that the endpoint environment variable is set to the correct host and port for your inference server. In this case, http://localhost:5000 is where our Flask application is running. ‘inference.’ $ export MINIO_NOTIFY_WEBHOOK_ENABLE_<YOURFUNCTIONNAME>=on 
$ export MINIO_NOTIFY_WEBHOOK_ENDPOINT_<YOURFUNCTIONNAME>=http://localhost:5000 Now, restart the MinIO Server using the command or if you’re starting the server for the first time you can also just use the command. For more information about re/starting the MinIO Server, check out the MinIO docs. Note: ALIAS should be replaced with the alias for your MinIO server deployment. For more information about how to set an alias or view existing aliases, check out . mc admin service restart ALIAS minio server the docs Finally, let’s add the bucket and event we want to be notified about. In our case, we want to be notified about ` events (creation of new objects) in our bucket. I made a brand new, empty bucket for this purpose titled so I will substitute that for ‘BUCKET.’ put` “detect-inference,” $ mc event add ALIAS/BUCKET arn:minio:sqs::<YOURFUNCTIONNAME>:webhook --event put You can check that you have configured the correct event type for the bucket notifications by verifying whether " " is outputted when you run this command: s3:ObjectCreated:* $ mc event ls local/detect-inference arn:minio:sqs::<YOURFUNCTIONNAME>:webhook For a more detailed explanation of publishing bucket events to a webhook, check out the . We are now ready to try object detection on a brand new image! docs Trying out our Inference System Here’s the new image (titled ‘1.png’) I want to conduct inference on: I drop the new image in my ‘detect-inference’ bucket: Almost instantly, I’m able to see the following results on my Flask server: $ python detection_server.py
 * Serving Flask app 'detection_server'
 * Debug mode: off
 * Running on http://127.0.0.1:5000
Press CTRL+C to quit

image 1/1 /var/folders/xf/q7x0z8sn5nvckccp1g0m1vpm0000gn/T/tmpo6jx3w8u/1.png: 448x736 2 SU-30s, 101.0ms
Speed: 4.1ms preprocess, 101.0ms inference, 5.8ms postprocess per image at shape (1, 3, 448, 736)
{'results': [[1927.78369140625, 627.7123413085938, 1995.090576171875, 715.3443603515625, 0.8142037987709045, 0.0], [1735.740234375, 477.2108154296875, 1809.181640625, 555.767578125, 0.7766116261482239, 0.0]]}
127.0.0.1 - - [14/Sep/2023 15:39:21] "POST / HTTP/1.1" 200 - Note that each detected bounding box in the results list is in the YOLO format [x1, y1, x2, y2, probability, class]. Here are the bounding boxes and predicted classes superimposed on the original image: Note: For production environments and/or large machine learning models, it’s a good idea to use an established model serving framework like PyTorch Serve or Triton Server to make inference more robust and reliable. If you’re interested in that, check out the previous post on . Optimizing AI Model Serving with MinIO and PyTorch Serve Concluding Thoughts We did it! We walked through how MinIO and CVAT came together to keep our collected image samples safe and available, as well as how to create our custom object detection dataset. Then, we trained our own custom YOLO model for our custom task. Lastly, with just over 50 lines of code, we put together an inference server using MinIO Bucket Notifications that could run a new image past our custom trained object detection model. Furthermore, for most mission critical applications of computer vision, it is best to conduct inference on the edge. Otherwise, said applications are vulnerable to the latency associated with uploading new data to the public cloud and waiting for an inference server in the cloud to get back with an answer– not to mention the risks of a faulty network connection. For this reason, a computer vision pipeline centered around MinIO as the data layer makes much more sense. Imagine a drone flying over an airfield, being able to capture, store, and use our trained model on new imagery with completely-on-board hardware and software. With a local deployment of MinIO server, the Bucket Notification-based inference system we built at the end of the post works perfectly for this scenario and countless others like it. If you have any questions join our or drop us a note at . We are here to help you. Slack Channel hello@min.io Also published . here

How to Train an Object Detection Model Using a Custom Dataset With MinIO and YOLO

MinIO এবং YOLO এর সাথে একটি কাস্টম ডেটাসেট ব্যবহার করে কীভাবে একটি অবজেক্ট সনাক্তকরণ মডেলকে প্রশিক্ষণ দেওয়া যায়

So trainieren Sie ein Objekterkennungsmodell mithilfe eines benutzerdefinierten Datensatzes mit MinIO und YOLO

Cómo entrenar un modelo de detección de objetos utilizando un conjunto de datos personalizado con MinIO y YOLO

Comment former un modèle de détection d'objets à l'aide d'un ensemble de données personnalisé avec MinIO et YOLO

मिनिओ और योलो के साथ कस्टम डेटासेट का उपयोग करके ऑब्जेक्ट डिटेक्शन मॉडल को कैसे प्रशिक्षित करें

MinIO と YOLO でカスタム データセットを使用して物体検出モデルをトレーニングする方法

MinIO 및 YOLO와 함께 사용자 정의 데이터 세트를 사용하여 객체 감지 모델을 훈련하는 방법

Como treinar um modelo de detecção de objetos usando um conjunto de dados personalizado com MinIO e YOLO

Как обучить модель обнаружения объектов с использованием пользовательского набора данных с помощью MinIO и YOLO

MinIO ve YOLO ile Özel Bir Veri Kümesi Kullanılarak Nesne Algılama Modeli Nasıl Eğitilir

Cách đào tạo mô hình phát hiện đối tượng bằng cách sử dụng bộ dữ liệu tùy chỉnh với MinIO và YOLO

如何使用 MinIO 和 YOLO 的自定义数据集训练对象检测模型

Philosophy has the term "simulacrum" - a copy of something that does not actually exist. An excellent example of a simulacrum is the current Russian propaganda glorification of its soldiers, which imitates the Soviet propaganda heroic epic of the Patriotic War. One of the pillars of Russian propaganda is the Great Patriotic War. This saga has programmed Russians since childhood into the desired system of thinking, where there is a great state that has defeated evil, and someone else is only jealous and wants to return the evil. Therefore, Russia believes it has the right to deal with evil as it pleases because the truth is on its side. This cult of war replaces the religion that unites all faiths. To support it, the Kremlin has romanticized World War II for years. Now, however, Moscow has begun to cultivate a new war, which is supposed to be something that will unite a new generation of Buryats, Dagestanis, or residents of St. Petersburg. Moscow wants to join Russians with what World War II was for Soviet citizens, even though war with Ukraine is invading. If you want to get deeper into the Russian-Ukraine war, I recommend you to watch . 10 Films About the War in Ukraine Russian propaganda creates a pantheon of heroes from the occupiers Let's use the example of the "Operation Z Heroes" project, which publishes the allegedly heroic deeds of Russian army soldiers during the invasion of Ukraine, to see how it works. The project itself is about real people. But there are a few buts: In the description of their feat, there is no data about the place of the event, the military unit, or the date when it happened. Everything is done so that journalists cannot verify the story's reliability. In most of the videos, there is no data about who the heroes of the operation are fighting. It would be strange to glorify the occupiers fighting in another country. Therefore, in the description of their actions, the Ukrainian army is referred to as "Nazis" and "militants" or simply "enemies" Some videos do not even mention the word "Ukraine". Most exploits are created according to the Soviet propaganda model and are clearly hyperbolized. At the same time, we are talking about a low-quality duplicate because all of these deeds have real prototypes in Soviet history. But the style of warfare has changed, and now the description of such deeds looks like a mockery of Soviet soldiers. Russian propaganda was preparing Russians for the war with Ukraine even in games. I recommend you to read my article . How Russian Games Propagated the War Concrete examples of how Russian propaganda creates heroes from occupants One of WWII's most promoted Soviet propaganda heroes is the pilot , who crawled to his positions for 18 days. According to legend, this was after his plane was shot down and his feet were shattered during the landing. Alexey Maresyev Now there is a new story about Alexey Likhanov. He is not a pilot, but according to the legend, he was wounded, and it took him 12 days to return to his position with two grenades in his hands. Why Alexey was holding grenades all this time is unknown, as is where it happened. It is also unknown how he could crawl deep into the defense during the advance of his troops to crawl back for 12 days. An equally famous example of the Soviet epic is the communications officer Matvey Putilov, who died in the battle for Stalingrad. According to legend, he clenched the cable with his teeth to ensure communication while surrounded. That's how he was found dead. The Russians decided to take quantity rather than quality because modern troops do not need cables to provide communications. Russian propaganda created Sergeant Yan Volkov, who, for some reason, was left alone with a radio communications station and came under mortar fire. He, wounded, continued tuning it under fire. As a result, Russian troops destroyed the enemy grouping. Hussein Gadaev allegedly did something similar. His communications battalion was caught in an unequal battle, and the lieutenant was giving orders for communications during it until he lost consciousness. In general, this story is more evidence of the competence of Hussein's deputies. Nevertheless, he is in the pantheon of new heroes. Evstafiy Anisimov is a hero of the Soviet Union because, during the liberation of Belarus from the German invaders, he demined a bridge. According to legend, he did so under constant enemy fire. Russian propaganda went further and created the story of Kirill Balagansky, who allegedly demined two bridges under bullets. This story did not mention Ukraine because the Russian propaganda decided not to emphasize that Kirill, unlike Anisimov, was an invader. On the contrary, the hero of the USSR, Pavel Khvostov, built a bridge across the Dnipro. Under the bullets, his group could erect a pontoon crossing and ensure the further advance of the Soviet troops. Khvostov himself was wounded during the construction of the crossing. The Russians have a new story about Lieutenant Colonel Andrey Garanin also commanded the erection of a pontoon crossing under enemy fire. It is not specified which enemy and during the crossing of which river. However, it is proudly noted that thanks to the crossing of the unnamed river, Russian troops continued their advance to provide humanitarian aid to an unknown settlement on the other bank. However, the inhabitants of this unknown village would not have needed humanitarian aid if the occupant Garanin and his team had not invaded another country. General examples of how Russian propaganda creates heroes from occupants The list of new "heroes" includes a bunch of stories about nurses who carried the wounded. For example, Arkady Smolnikov saved a thousand wounded after the "treacherous Ukrainians" struck a "Russian-controlled airfield" (in common parlance, an occupied airfield). These stories duplicate the many stories of "Red Army" medics with several times inflated numbers of evacuees. In all cases, the main thing is the event's location is not established. Nobody even shows a video of the evacuation. In general, there are many propaganda clichés on the list: soldiers who "first noticed the enemy" (during the offensive), pilots who covered others with their aircraft (like the ), soldiers who shot down "a flock of bayraktars” and soldiers repelled volleys of MLRS "Smerch.” When propagandists ran out of imagination, they added to the list people who died or "almost died" (Warrant Officer Arsalan Munkuyev almost died). Soviet pilot Viktor Talalikhin Conclusion Russian propaganda easily devalues World War II heroes by rewriting their histories. As mentioned above, the war in the Russian Federation is a religion. Even if a Russian does not actively support the war (being an "atheist"), his morals are still shaped by this religion. He will celebrate war holidays and live the ethics of war, where there are evil and martyrs. Just now, Moscow is giving birth to a new cult with new heroes, a "new covenant" of war, a kind of "Putin's gospel." In such an information flow, Russians will not be able to draw conclusions from the war because it is being romanticized again. Thus, this war will not discourage Russians from continuing to fight; after Ukraine, Russia will prepare for a new war. And not necessarily against Ukraine, because of the very essence of the war as such. It is easy to create an enemy in this paradigm of thinking. So the task of Ukraine and its allies is to ensure that Russia is deprived of resources and physically unable to continue the war, which is exactly what is happening now. I recommend reading and . The Evolution of Warfare in Ukraine: Three New Features Their Offices in Mariupol and Irpen Were Destroyed by Russians

How Russia is Creating a New "Great Patriotic War (WWII)" With Ukraine

Do the Russians want war? Obviously, they do. According to the Russian Public Opinion Research Center, 74% of Russians support the war in Ukraine. For 76 years, Russians have carefully preserved the memory of the horrors of World War II, and suddenly, in a united rush, they are ready to kill and die again. There are many reasons why this is so, and the Internet has already picked them apart. But especially impressive is the correlation between the degree of support for the war and television viewing time. The fact of influence of the blue screen on the human mind is clear. By the way, I recommend you read my other article, “ ” 10 Films About the War in Ukraine I don't consider game developers part of some plan to turn Russia into a fascist state. But they created their product in the tone of what was going on in the country to satisfy the consumer's demand. The game development industry supported the processes that led with the consent of the Russians. Russia to attack Ukraine How Russian games started war propaganda? Until about 2003, the Russian game market was no different from any other. There were various games, genres, and dreams of success in Europe and the USA in this market. But in 2003, two critical things happened. In March 2003, America started a war in Iraq, which led to a sharp deterioration in U.S.-Russian relations. And in May of the same year, the famous game Blitzkrieg, based on World War II, was released. At that moment, two things became evident to the developers: The income level in Russia grew, and the profitability of projects focusing specifically on the Russian market increased. A game about World War II was the most in-demand thing that could be made for the Russian market. Blitzkrieg was a game with a story about the cumulative heroic feat of the Russian people who saved the whole world. The division of the Russian game development industry In the following years, U.S.-Russian relations continued to deteriorate. Anti-American and anti-European sentiment grew, while Russia's connection to the modern world weakened. At this point, two conventional directions emerged in the Russian game development industry, which influenced the future triumph of fascism. One direction actively exploited the audience's homesickness for the past. Russians wanted games in which they could see epic battles and take part in them personally. Numerous real-time strategies, military-themed games, tactics, and military equipment simulators emerged. All this helped cultivate what Archpriest Georgy Mitrofanov called "victoriousness" in 2005. That is hypertrophied pride in the victory of the USSR over Nazi Germany. The Russians began to be dominated by a sense of themselves as the heir to a victorious nation. A nation that is responsible for the world and deserves the gratitude of the world. In the same year, 2005, xenophobic marches with the slogans "Russia for Russians" began in Russia. During this period, the Russians added to the "victorious" history of their people those war episodes that have a reputation for being tragic or even shameful. Thus, in 2005 the game "Alpha: AntiTerror" and its supplement "Men's Work" came out. These are tactical strategies where the player takes control of a detachment of Russian special forces and destroys the enemy in the Afghan and Chechen wars. Two games at once, "The 9th Company" and "The Truth about 9th Company," were devoted to a battle in which a small group of Soviet paratroopers defeated several hundred Afghan mujahideen. I wonder how Russians can appeal to the Afghan war, which they, according to polls, consider tragic and unnecessary. The Russians' belief in their innate heroism leads them to seek or create evidence of this heroism constantly. Hence the love for stories in the style of "300 Spartans," "9th Company," and "Panfilov's 28 Men". Not surprisingly, not even a few days into the , the Russians started spreading the legend of the "200 Spartans of Gostomel," who allegedly defended the airfield heroically from 1,000 "Ukronazis". war with Ukraine Russian game development industry propagandized that Russia is alone against the whole world In those years, when Ukraine had the Orange Revolution and the Baltic states at the very borders of Russia "asked" to join NATO, there was a growing feeling in Russia that enemies surrounded the country. The second direction of the Russian game development industry was responding to these feelings. There are several examples of such games: "Hammer & Sickle" in 2005, "Marauder" in 2009, "Confrontation: Peace Enforcement" in 2009, and "Peacemaker" in 2009. In all these games, invincible Russians or Soviet spies fight American spies, soldiers, and, of course, the NATO. "Confrontation: Peace Enforcement" is a clear example of conscious propaganda. The game's plot tells the story of how events would unfold if Georgia tried to take back South Ossetia. The developers have even put a cartoon of Mikheil Saakashvili on the game disc, showing how this “coward” starts chewing his tie at the mere mention of Russian tanks. In all these games, the player is always the victim, defending himself against the enemy and never attacking himself. Better yet, the player defends some third party too weak to defend himself. Thanks to Russian propaganda, "never again" has become "we can do it again" By about 2012, the Russian government was actively pumping propaganda into its citizens and promoting two important ideas: Russia justified its right to influence the world because it had saved the world from the nazis. The significance of the allied contribution to the victory was devalued. In addition, Russia again put on the dress of the victim and accused the U.S. of being the one who denied the USSR's role in the victory. Victory Day (9 May) in Russia was completely changed to Day of Remembrance and Sorrow, and the slogan "never again" was finally turned into "we can do it again". Since then, 9 May has been fully dedicated to demonstrating the titanic power of Russia equal to that of the USSR. The number of Russian marches increased, and then Russia directly "shone" on the world stage: Crimea, Donbas, and Syria. What was going on with Russian games during this period In 2010, the game "Confrontation. 3D. Reloading" was released. In the story, because of the U.S. actions in Afghanistan, there was a rampant growth of drug trafficking, and Russian soldiers are forced to stop the lawlessness. Shortly after the occupation of Crimea, 1C began developing the game "Caliber". At first, the game was called "Polite People". From my point of view, this was a powerful message to the target audience in Russia. "Remember how cool our soldiers took Crimea from the Ukrainians? Want to be just as cool? Play our game!" In 2017 the game Escape from Tarkov came out. Of course, it is primarily a good game, not propaganda. Nevertheless, the developers kept their nose to the grindstone: the story is full of enemies, the British are to blame, and the Russians oppose them all. In 2009, Russia began to develop the latest Armata tank. The tank, as always, had no analogs and was much better than the Americans. It was a formidable confirmation that Russia could indeed repeat. And in 2014, a year before the Armata was introduced to the general public, the Russian corporation Uralvagonzavod (a Russian company for the production of military equipment) dropped off money to the developer of the Armored Warfare tank simulator. Then, the companies announced cooperation, and the game began to publicize the formidable power of the Russian armed forces. Games played a huge role in exaggerating the power of the Russian army. Can Russia propagandize war ideas with the help of the metaverse? While Meta is fine-tuning its world's fastest supercomputer to create its meta world, China and Russia are looking for ways to use metaverse for their interests. For example, the Chinese government is going to take full state control of the Chinese version of the meta world. Russian authorities also want to be in the trend. The risks of using the for propaganda by Russia are significant. For example, implementing trade operations within a metaverse using virtual money can be used for financing terrorism or armed conflicts. In addition, Russia can use illegal political activity in the form of agitations, protests, and propaganda in such a space. Another danger is the theft of users' identities hiding behind virtual avatars. metaverse I am sure Russia will invest in propagating its crazy ideas in the metaverse. Likewise, any other country, religion, or organization can propagate its ideas. The only question is how to keep people in the metaverse away from propaganda. Conclusion Sometime in 2008, the Russian government became aware of the existence of the game development industry and began to use it for its purposes. And game development industry began to realize what kind of country it was in. A part of the game development industry became horrified and stopped making games that had anything to do with reality and Russia. Another part liked what they saw, and the developers fell into outright propaganda. In the end, the monster of Russian fascism grew up in two main senses: A sense of its power, greatness, and exceptionalism A sense of continuous threat from outside and the belief that the whole world is against the Russians Even without purpose other than to do what they love and make money, Russian game developers have fueled this Russian propaganda on par with other art and culture. Most of all, I am concerned about the future of Russian propaganda in the metaverse and the ways to stop it. I also recommend you to read: The 7 Craziest Financial Stories of 2021: Investing in Nothing VR Technology and Metaverse Learning: Advantages and Disadvantages The 19th-Century Dot-Com Bubble: How a Scotsman Sold a Fake Country Real Estate in the Metaverse is Booming: is it Really Such a Crazy Idea? Web3 for Beginners: Why Web3 Will Change Everything?

How Russian Games Propagated the War

Cómo los juegos rusos propagaron la guerra

Comment les jeux russes ont propagé la guerre

कैसे रूसी खेलों ने युद्ध का प्रचार किया

ロシアのゲームが戦争を広めた方法

Como os jogos russos propagaram a guerra

Trò chơi Nga tuyên truyền chiến tranh như thế nào

俄罗斯游戏如何传播战争

It is important to realize that we have the ability to manufacture our own fate when we want to. We can… proceed when things look bad, or we can find plenty of reasons to quit if we don’t want to go forward. After numerous terrorism incidents in the 1970s, the United States realized its armed forces had a blind spot when it came to counterterrorism. Green Beret , who had served alongside the British Army’s counter-terrorism unit the Special Air Service (also known as “ ”) in Malaysia, had been pushing for such a group since the 1960s. Colonel Charlie Beckwith SAS Now that terrorism was a proven threat killing Americans and the opportunity to be proactive had passed, the U.S. Army decided it was a good time to commission its own SAS-like force — namely, . Delta Force It is in the lead-up to the formation of Delta Force that we meet Eric Haney. You don’t always have to know what you are getting into Haney grew up in the mountains of northern Georgia during the 1950s and remembers, as a young boy, how exciting it was when electricity and indoor plumbing were added to his family home. His parents had never entered high school, so the fact that he graduated high school was more than they could have asked for. Following in the footsteps of friends and family, Haney joined the Army straight out of high school. Though we may not have been scholars, we did know how to go into the military. I had grown up listening to the war stories and tales of my family and friends and I was determined to join up just as soon as I was able. I enlisted in the Army in the spring of 1970, while still in high school, with a reporting date immediately after graduation. I fell in love with the Army as soon as I met her. I became a professional soldier, and that is what I will be until I die. The military is a profession that brands itself on the soul and causes you forever after to view the world and all human endeavor through a unique set of mental filters. The more profound and intense the experience, the hotter the brand, and the deeper it is plunged into you. I was seared to the core of my being. About the time Colonel Beckwith was starting to recruit for Delta Force, Haney had just been promoted to sergeant first class in the Army Rangers. Haney enjoyed his job — and he was good at it. But after eight years of near-continuous deployment, he was fretting that this recent promotion meant an imminent re-assignment as a state-side instructor training subsequent Ranger generations. In 1978, Haney had just returned to Georgia after a month in the jungles of Panama when he stumbled into the opportunity to try out for one of the first Delta Force squadrons. Beckwith sent one of his men to offer Haney the opportunity in person. He had my personnel records open on the table in front of him and he glanced at them occasionally as we talked about my career, about the units I had been in and the assignments I had held to that point. He told me this was a chance to be a charter member of a unit that would be unique in the American military — the nation’s first unit dedicated to fighting international terrorism. The prerequisites to try out were: Minimum age of twenty-two. Minimum time in service of four years and two months. Minimum rank of staff sergeant. Pass a 100 meter swim test while wearing boots and fatigues, and pass the Ranger/Special Forces PT test. Have a minimum score of 110 on the Army general aptitude test, no court-martial convictions, and no record of recurring disciplinary problems. About the only other thing Grimes told me was that if I was accepted, I could expect hard work, plenty of danger, and no recognition. Given his fear that he would get stuck as a state-side trainer, Haney wasted no time — he signed up for the tryouts on the spot. I had almost no idea of what to expect on the morning of 13 September 1978, when I loaded my pickup, kissed my family goodbye, and set out on the five-hour drive up I-95 from Hunter Army Airfield in Savannah, Georgia, for Fort Bragg, North Carolina — and points unknown. I merely reminded myself that the future is always perfect. And wished it to be so. Perseverance is great… The tryouts were a continuous string of examples showcasing the often-said-not-often-followed truism that hard work and persistence are the bare minimum qualities required to achieve anything meaningful. Haney participated in run-of-the-mill running tests, an 18-mile hike with 40 pounds on his back, solo navigation drills across mountains and valleys where the destination was never known, and it all culminated with a 40-mile mountain hike that started at 2 am. And to top it all off, the testing officers were an everlasting source of demotivation in the field — “You can just quit now if you’d like” was an often shared phrase. And what a beautiful metaphor for life that is. Your destination is never truly known, it requires unending effort to achieve things worth achieving, and along the way you can easily find sources of negativity telling you to voluntarily remove yourself from the challenge. However, Haney was not caught up in the grandeur of the end goal and how to get there most efficiently. That mindset doesn’t work when you don’t know where the end actually is. Instead, he was steadfastly focused on doing well along the way and trusting he would get where he was going eventually — he just had to trust in himself along the way. I would just keep my mouth shut, my eyes and ears open, and respond to whatever came up. It’s the system I’d always used in new situations, and so far it had served me well. Additionally, as he attacked what was in front of him he pushed away the temptation to look around and worry about comparing himself to other applicants. Some were faster. Some were stronger. . But those things didn’t matter — every time he was close to quitting, he would push himself a little further, get his second wind and keep on trucking. Some seemed to mind the pain less than he did As the truck rumbled off, I looked at the other group, but they were still sitting there. I quickly wondered which of us was going where, and just as quickly dismissed the thought. The men in the other group weren’t my concern, and as for me, I’d know the destination when I got there. And because of this, he got to the destination while many others did not. … But don’t forget to prepare And as you think through how impossible these challenges sound, you are met with the realization of how quickly you would fail participating in the same. But then you have to be honest with yourself — you wouldn’t just fail, you would, in fact, die. You would get lost in the mountains. You would fall off a cliff. You would try to make the 40-miles and breakdown before you ever got to the end. And this highlights another useful takeaway — unending resolve is not enough to ensure success. You must also be competent. All the perseverance you can muster today won’t do anything for you if you haven’t lived a life up to today that has prepared you for the task at hand — . who you are today is a lagging indicator of who you have been This means that success can only come after persistent preparation. The 163 men selected for the tryouts had lived lives that prepared them to take the unique, once-in-a-lifetime Delta Force opportunity and try. Without preparation, they could not even hope to try. This is not unique to the military — consistently shares the idea that there are only a few opportunities in life that define you. To take advantage of them, you prepare. Charlie Munger Spend each day trying to be a little wiser than you were when you woke up. Discharge your duties faithfully and well. Step by step you get ahead, but not necessarily in fast spurts. But you build discipline by preparing for fast spurts. - Charlie Munger Just because you don’t achieve your original goal does not mean you are a failure However, even with preparation, you will make mistakes. Haney did just that on the final 40-mile hike. He got lost. He went 7 miles off course in the wrong direction. And had to backtrack. And then keep going. He turned a 40-mile hike into more than a 50-mile hike, but he finished. After 18 hours of hiking, he made it to the finish line despite the setback. This is another takeaway — it is rare that failure is a permanent state. After failure, you have the option to keep going. You can treat failure as only a stop on your journey. . Many people treat getting knocked down as a reason to stay down. Like Haney, we don’t have to. You don’t have to let it be the destination What was the lesson here? Simple. Don’t quit. Never quit no matter what. Keep going until someone tells you to sit down. Keep going as long as you’re able to move, no matter how poorly you think you may be doing. Just don’t quit. And it is important in life that you don’t define failure as failure to achieve only one specific goal. You can show up prepared and still not cross a specific finish line. 163 men participated in the Delta Force trials, but only 12 were ultimately accepted into the program. Those 151 men were not failures — not even close. They just had to accept that this Delta Force finish line wasn’t one they were going to cross . and instead start progressing towards the next Previously published at https://newsletter.butwhatfor.com/p/perseverance-is-great-but-dont-forget

If You Want To Achieve A Goal Show Up Prepared

Motivation A time-series represents the sequence of a variable recorded with time over regular intervals of time. There are various applications of time series like Stock Market Prediction (with daily and hourly data), Annual Sales/Revenue, Seasonal Temperatures changes, monthly Cloud Infrastructure cost and second level prediction of Python has 2 libraries StatsModels and Pyramid that helps to build forecasting models and predict values at a future time. As both these libraries provide forecasts for similar models namely, AR (Auto-Regressive)MA (Moving Average)ARIMA (Auto-Regressive Integrated Moving Average)SARIMA (Seasonal Auto-Regressive Integrated Moving Average) it will be interesting to see the features provided by each of the libraries along with the pros and cons of using each of them. In this blog, I try to summarise the functionalities of both of these libraries by demonstrating the Number of Active Cases for Covid-19 for any Indian state. We keep our scope limited to univariate time series analysis. DataSet The scope of the data is fro 8th March-1st May where data from 8th March-14th April has been used for training and 15th March-1st May has been used for validation and prediction. Here, we have demonstrated using state Maharashtra, however all states data and modeling can be done using the processed data from . https://github.com/sharmi1206/covid-19-analysis STATSMODEL It primarily comprises of three main modules: – Comprises of functions useful for univariate time series analysis – autoregressive models (AR), vector autoregressive models (VAR), and univariate autoregressive moving average models (ARMA), along with Non-linear models like dynamic regression and autoregression. In addition, it also provides autocorrelation, partial autocorrelation function, periodogram, and other statistical properties related to ARMA, and methods that process moving average lag-polynomials. tsa –  Equipped with functionality to model observations at any time instant with unobserved state vectors and irregular components. It has added flexibility to allow estimation with missing observations, forecasting, impulse response functions, etc. This flexibility helps it to differentiate from tsa by estimating additive and multiplicative seasonal effects on models (both on univariate and multi-variate), as well as arbitrary trend polynomials. statespace -Allows simultaneous modeling and analyzing multiple time series. vector_ar ARIMA MODELS An ARIMA model is composed of 3 constituent units which are : : . This part explores any dependent relationship between an observation and some number of lagged variables. AR Autoregression : . This part aims to make time-series stationery by subtracting or differencing an observation from observation at the previous time step of the same time series. I Integrated : This part explores the relationship between an observation and a residual error by application of moving average to lagged observations, with any given time window. MA Moving Average . In order to design a model (say ARIMA) by applying all of the above stated procedures we need to enable all parameters. We can also enable partial parameters and select to choose either AR or MA model. The parameters can be summarized as follows: : The number of lag observations included in the model, also called the lag order. p : The number of times that the raw observations are differenced also called the degree of difference. d : The size or the order of the moving average window. q To start with and in order to have different flavours of ARIMA model, the library imports are: statsmodels.tsa.ar_model AR statsmodels.tsa.arima_model ARIMA statsmodels.tsa.statespace.sarimax SARIMAX statsmodels.graphics.tsaplots plot_acf,plot_pacf from import from import from import from import MANUAL MODEL SELECTION ARIMA The following diagram illustrates the mechanism of using AR/MA/ARIMA/SARIMA by self choosing the model orders (p, q, d) and self tuning it. AR Model This Auto-Regressive model estimates a dependent variable at any timing instant based on some number of lagged observations. The variables can be summarized as follows and formulated by the below-given equation: Yt Y(t-1): Lag 1 of the time series β: Coefficient of lag1 α: Intercept term Dependence of Lag in AR Estimation, where β and α are model estimates Number of Active cases for Covid-19 can be formulated using AR model as follows. Here we put p, i.e. number of lag observations as 2, while d and q are set to 0. model_ar = ARIMA(trainActiveCases, order=( , , ))  
model_ar_fit = model_ar.fit()  

prediction_ar=model_ar_fit.forecast(len(validActiveCases))[ ]

print(model_ar_fit.summary()) residuals = pd.DataFrame(model_ar_fit.resid)
residuals.plot()
plt.show()
residuals.plot(kind= )
plt.show()
print(residuals.describe()) 2 0 0 0 # plot residual errors 'kde' The summary and the predicted output obtained from the model are illustrated below: AR Moel Summary Predicted vs Actual AR MA Model The MA model also estimates a dependent variable Yt based on the lagged forecast errors. MA Estimation with Lagged Errors where the error terms are the errors introduced from respective lags. The errors Yt and Y(t-1) are the errors from the following equations : Number of Active cases for Covid-19 can be formulated using MA model as follows. Here we put ma (moving average window), q as 2, while d and p are set to 0. model_ma = ARIMA(trainActiveCases, order=( , , ))  
model_ma_fit = model_ma.fit()  

prediction_ma=model_ma_fit.forecast(len(validActiveCases))[ ]

print(model_ma_fit.summary()) residuals = pd.DataFrame(model_ma_fit.resid)
residuals.plot()
plt.show()
residuals.plot(kind= )
plt.show()
print(residuals.describe())

model_scores.append(np.sqrt(mean_squared_error(validActiveCases,prediction_ma)))
print( ,np.sqrt(mean_squared_error(validActiveCases,prediction_ma))) 0 0 2 0 # plot residual errors 'kde' "Root Mean Square Error for MA Model: " The summary and the predicted output obtained from the model are illustrated below: MA Model Summary Predicted vs Actual MA ARIMA Model An ARIMA model comes along with both the flavours of AR and MA model where the time series is differenced at least once to make it stationary. The equation can be formulated as follows, where we see the term β coming from AR and ε–the error terms coming from MA model . ARIMA – Differencing + Combination of Parameters from AR and MA model Number of Active cases for Covid-19 can be formulated using MA model as follows. Here we put q i.e the ma (moving average window), d (order of differencing ) and p (number of lag observations) are set to 1. The code snippet for this is : model_arima = ARIMA(trainActiveCases, order=( , , ))  
model_arima_fit = model_arima.fit()  


prediction_arima=model_arima_fit.forecast(len(validActiveCases))[ ]
valid_ml[ ]=list(np.exp(prediction_arima))

print(model_arima_fit.summary()) residuals = pd.DataFrame(model_arima_fit.resid)
residuals.plot()
plt.show()
residuals.plot(kind= )
plt.show()
print(residuals.describe() 1 1 1 0 "ARIMA Model Prediction" # plot residual errors 'kde' Predicted vs Actual ARIMA We can see it’s better to use ARIMA than AR or MA model, as this is the model that closely resembles the actual outcome. SARIMAX Model SARIMAX model is built by extending the ARIMA model, discussed in our previous section. In addition to terms and terms, there are four seasonal elements that are not part of ARIMA that must be configured; they are: AR, I MA : Seasonal autoregressive order. P : Seasonal difference order. D : Seasonal moving average order. Q : The number of time steps for a single seasonal period. m Number of Active cases for Covid-19 can be formulated using the model as follows. Here we put ma (moving average window), d (order of differencing ) and p (number of lag observations) are set to 1. We also enforce seasonal order as (0,0,0,12) to study its behavior, even though Covid-19 does not show any seasonality, evident from research and daily reports. SARIMAX The code snippet for SARIMAX (without exogenous variable) can be given us: model_sarima = sm.tsa.statespace.SARIMAX(trainActiveCases, order=( , , ), seasonal_order=( , , , ),
                                 enforce_stationarity= , enforce_invertibility= )
model_sarima_fit = model_sarima.fit()  


prediction_sarima=model_sarima_fit.forecast(np.shape(validActiveCases)[ ])
valid_ml[ ]=list(np.exp(prediction_sarima))

print(model_sarima_fit.summary()) residuals = pd.DataFrame(model_sarima_fit.resid)
residuals.plot()
plt.show()
residuals.plot(kind= )
plt.show()
print(residuals.describe()) 1 1 1 0 0 0 0 False False 0 "SARIMA Model Prediction" # plot residual errors 'kde' The model summary and the predicted outcome are illustrated from the below plots. SARIMAX Model Summary Predicted vs Actual SARIMA In order to plot the trained, validation and predicted data under proper dates, we use the following code snippet. plt.figure(figsize=( , ))

index= pd.date_range(start=train_dates[ ], periods=len(train_dates), freq= )
valid_index = pd.date_range(start=valid_dates[ ], periods=len(valid_dates), freq= )

train_active =  pd.Series(train_ml[ ].values, index)
valid_active =  pd.Series(valid_ml[ ].values, valid_index)
pred_active =  pd.Series(prediction_ar, valid_index)

f, ax = plt.subplots( , , figsize=( , ))
plt.plot(train_active, marker= ,color= ,label = )
plt.plot(valid_active, marker= ,color= ,label = )
plt.plot(pred_active, marker= ,color= ,label = )

plt.legend()
plt.xlabel( )
plt.ylabel( )
plt.title( + stateName) 10 5 0 'D' 0 'D' 'Active Cases' 'Active Cases' 1 1 12 10 'o' 'blue' "Train Data Set" 'o' 'green' "Valid Data Set" 'o' 'red' "Predicted ARIMA/SARIMAX" "Date Time" 'Active Cases' "Active Cases ARIMA/SARIMAX Model Forecasting for state " Auto-Correlation Plots Differencing helps to make time-series stationary. In order to find a right order of differencing, we use the ACF plot and see that it reaches zero, fast and with a fairly static mean. With Covid-19 data for state Maharashtra, the ACF plot and the code snippet are given below. fig, axes = plt.subplots( , , sharex= , figsize=( , ))
axes[ , ].plot(df_per_State_features[ ]); axes[ , ].set_title( )
plot_acf(df_per_State_features[ ], ax=axes[ , ]) axes[ , ].plot(df_per_State_features[ ].diff()); axes[ , ].set_title( )
plot_acf(df_per_State_features[ ].diff().dropna(), ax=axes[ , ]) axes[ , ].plot(df_per_State_features[ ].diff().diff()); axes[ , ].set_title( )
plot_acf(df_per_State_features[ ].diff().diff().dropna(), ax=axes[ , ])

plt.show() # Original Series 3 2 True 10 5 0 0 'Active Cases' 0 0 'Original Series' 'Active Cases' 0 1 # 1st Differencing 1 0 'Active Cases' 1 0 '1st Order Differencing' 'Active Cases' 1 1 # 2nd Differencing 2 0 'Active Cases' 2 0 '2nd Order Differencing' 'Active Cases' 2 1 AUTO ARIMA MODELS This library automatically discovers the optimal order for an ARIMA model with stepwise execution of hyperparameters and parallel fitting of models. This objective of this library (auto_arima) is to identify the most optimal parameters for an ARIMA/SARIMA and return a fitted ARIMA model. It does not depend on the , but instead, it conducts differencing tests (i.e., Kwiatkowski–Phillips–Schmidt–Shin, Augmented Dickey-Fuller or Phillips–Perron) on its own to determine the order of differencing. PACF/Auto-Correlation (manual computation of differencing) One other difference from Stats Model’s Vanilla ARIMA is that in case the model does not converge, it raises ValueError to signify the re-evaluation of stationarity-inducing measures. For seasonality, SARIMA identifies the optimal P and Q hyperparameters after conducting the Canova-Hansen to determine the optimal order of seasonal differencing , so that the auto-regressive/moving average portions of the seasonal model are defined by start_P, start_Q, max_P, max_Q D The model is fitted within ranges of defined start_p order (or number of time lags for AR), max_p, start_q (order of the moving average MA), max_q ranges. Hence it can be formulated as : (p, q, d)x(P, Q, D) The best model is obtained by optimizing ( – Akaike Information Criterion, – Corrected Akaike Information Criterion, – Bayesian Information Criterion, – Hannan-Quinn Information Criterion, or “out of bag”–for validation scoring–respectively to a minimal value. information_criterion AIC AICC BIC HQIC It also operates with variables (just like state space methods/models) for predicting added features in the regression operation. exogenous Summary of AR with Auto-ARIMA The following code and figure depicts AR model with Auto ARIMA with start_p=0, start_q=2 (by default), max_p=5, max_q=0 model_ar= auto_arima(trainActiveCases,trace= , error_action= , start_p= ,start_q= ,max_p= ,max_q= ,
                   suppress_warnings= ,stepwise= ,seasonal= )
model_ar_fit = model_ar.fit(trainActiveCases)
prediction_ar=model_ar_fit.predict(len(validActiveCases))

print(model_ar_fit.summary())
plt.figure(figsize=( , )) residuals = pd.DataFrame(model_ar_fit.resid())
model_ar_fit.plot_diagnostics()
plt.show()
residuals.plot(kind= )
plt.show() True 'ignore' 0 0 5 0 True False False 10 10 # plot residual errors 'kde' Predicted vs Actual AR Auto-ARIMA Now let's look at the Moving Average model by changing (set to 5). Here's the code snippet for it. start_p (AutoRegressive Moving part - set to 0), start_q, max_p, max_q model_ma= auto_arima(trainActiveCases,trace= , error_action= , start_p= ,start_q= ,max_p= ,max_q= ,
                   suppress_warnings= ,stepwise= ,seasonal= )
model_ma_fit = model_ma.fit(trainActiveCases)
prediction_ma=model_ma_fit.predict(len(validActiveCases))

print(model_ma_fit.summary())
plt.figure(figsize=( , )) residuals = pd.DataFrame(model_ma_fit.resid())
model_ma_fit.plot_diagnostics()
plt.show()
residuals.plot(kind= )
plt.show() True 'ignore' 0 0 0 5 True False False 10 10 # plot residual errors 'kde' The hyper-parameter tuning for evaluating the lowest and best AIC with the residual components are given below. Summary of MA with Auto-ARIMA Predicted vs Actual MA Auto-ARIMA The Auto ARIMA is then used to model ARIMA by using the following parameters ------------------ as shown in the below code snippet. (start_p=1,start_q=1,max_p=5,max_q=5) model_arima= auto_arima(trainActiveCases,trace= , error_action= , start_p= ,start_q= ,max_p= ,max_q= ,
                   suppress_warnings= ,stepwise= ,seasonal= )
model_arima_fit = model_arima.fit(trainActiveCases)
prediction_arima=model_arima_fit.predict(len(validActiveCases))

print(model_arima_fit.summary())
plt.figure(figsize=( , )) residuals = pd.DataFrame(model_arima_fit.resid())
model_arima_fit.plot_diagnostics()
plt.show()
residuals.plot(kind= )
plt.show() True 'ignore' 1 1 5 5 True False False 10 10 # plot residual errors 'kde' Summary of ARIMA with Auto-ARIMA Predicted vs Actual Auto-ARIMA AUTO SARIMA MODEL The SARIMA model is tuned with to find the best model with the lowest AIC. The code snippet is given below, which does parameter tuning for Here and are tuned with values (1,0,1), (1,0,0),  (0,0,1)  and 12 respectively. hyper-parameters (p, q, d)x(P, Q, D) (P, Q, D) (m) The parameters are initialized as , . even , takes start_p=0,start_q=2,max_p=5,max_q=5,m=12 start_q set to 0 default value of 2. model_sarima= auto_arima(trainActiveCases,trace= , error_action= , 
                         start_p= ,start_q= ,max_p= ,max_q= ,m= ,
                   suppress_warnings= ,stepwise= ,seasonal= )
model_sarima_fit = model_sarima.fit(trainActiveCases)
prediction_sarima=model_sarima_fit.predict(len(validActiveCases))

print(model_sarima_fit.summary())
plt.figure(figsize=( , )) residuals = pd.DataFrame(model_sarima_fit.resid())
model_sarima_fit.plot_diagnostics()
plt.show()
residuals.plot(kind= )
plt.show() True 'ignore' 0 2 5 5 12 True True True 10 10 # plot residual errors 'kde' Summary of SARIMA with Auto-SARIMA Predicted vs Actual Auto-SARIMA Seasonal Differencing SARIMA We also try to explore seasonality with Covid-19 dataset by applying regular differencing, where the values are differenced from the previous season. In the equation of SARIMA(p,d,q)x(P,D,Q): P denotes Seasonal AR, D denote Seasonal Order of seasonal differencing and Q denotes Seasonal Moving Average, x is the frequency of the time series. If a model has defined seasonal patterns, we set D=1 for a given frequency ‘x’. For a SARIMA model, we keep the total differencing + D’ <=2 SELECTION OF BEST MODEL with AUTO-ARIMA The following figure illustrates how to select the best Auto ARIMA Model, compute its error metrics for any input time-series data. Comparative Study of RMSE Conclusion and Future Work More work needs to be done by studying the impact of the number of people quarantined, hospital beds, icu units, ventilators, rate of tests done by using Vector Autoregressive Moving-Average with eXogenous regressors (VARMAX). Evaluate the model performance by (multiplicative models) and the two different modules of StatsModel. sm.tsa.statespace.ExponentialSmoothing sm.tsa.ExponentialSmoothing (linear model), i.e. A detailed study of using different smoothing techniques with latency and accuracy to analyze the tradeoff between the two. Evaluate confidence intervals for forecasts, based on an assumption of Gaussian errors as provided by Analyzing the performance impact by concentrating initial values out of the objective function, particularly in situations when the seasonal periodicity is large. sm.tsa.statespace.ExponentialSmoothing With more number of COVID-19 cases evolving, the right order of differencing (if needed multiple orders of differencing) needs to be computed to arrive at a near-stationary series (ACF plot nearing zero). Here we have only computed RMSE, but an ideal situation all error metrics Mean Absolute Percentage Error (MAPE), Mean Error (ME), Mean Absolute Error (MAE), Mean Percentage Error (MPE), Root Mean Squared Error (RMSE), Lag 1 Autocorrelation of Error (ACF1), Correlation between the Actual and the Forecast (core), Min-Max Error (min-max) needs to be computed. The complete source code is available at https://github.com/sharmi1206/covid-19-analysis References https://www.statsmodels.org/dev/user-guide.html#time-series-analysis http://alkaline-ml.com/pmdarima/0.9.0/modules/generated/pyramid.arima.auto_arima.html https://www.machinelearningplus.com/time-series/arima-model-time-series-forecasting-python/ https://machinelearningmastery.com/how-to-grid-search-sarima-model-hyperparameters-for-time-series-forecasting-in-python/https://otexts.com/fpp2/arima-r.html

Basic Understanding of ARIMA/SARIMA vs Auto ARIMA/SARIMA using Covid-19 Data Predictions

is a decentralized prediction market protocol built upon Ethereum which allows users to speculate on anything. Prediction markets are a sector ripe for disruption from blockchain technology; however, scalability problems and broader adoption remain as potential growth inhibitors when it comes to a project like Augur. Augur Reputation (REP) is the that serves as the backbone of the Augur ecosystem. Those who hold REP stake their tokens during the validation stage of prediction’s outcome. If a REP holder disagrees with the outcome reported, they can stake REP to challenge the result. These “reporters” are essentially declaring, with their REP, which market outcome actually matches the real-world outcome. REP can thus be viewed as a whose purpose is the assignment of value to the trustworthiness of reporters (or prospective reported-outcome-challengers) in a prediction market. Simply put, REP is only used for reporting and disputing event outcomes within Augur, while Ethereum is the currency actually traded on the protocol with the addition of in the upcoming upgrade. ERC-20 token utility token DAI v2 There are essentially innumerable use cases for a decentralized prediction market platform, which opens up the universe of questions that can be asked and reduces the regulatory worry that might come with a more centralized solution. Given the lack of other viable challengers to Augur in the crypto space, Augur is poised to capture this potential market though it will take time to see if it can garner any significant user attention and maintain steady growth. Technology Augur did not reinvent the wheel when it comes to blockchain technology; rather, it represents a smartly-coded, meticulous application of extant technologies—namely, Ethereum’s smart contracts—in order to facilitate the decentralized creation of prediction markets. While it came with no massive innovations in privacy or consensus mechanisms, Augur does represent a clean and practical application of , which is to be lauded. smart contract technology All actions within the Augur protocol such as matching and routing orders, processing trades, and resolving markets are all operated and run on the Ethereum blockchain itself through a set of smart contracts. Miners running Ethereum nodes route, match, and process orders on the Augur protocol. Economics The target utility of the REP token disincentivizes and instead encourages active participation in the prediction markets. This incentive helps drive Augur’s purpose and the main benefit of using a prediction market —the promotion of the wisdom of the crowds. That said, this same limited utility necessarily lowers the prospective value of the REP token, as it is not needed for the betting portion of market participation. One could reasonably argue that these economic incentives and the overall market design of the Augur protocol were orchestrated to maximize the efficiency of and participation in the various prediction markets, even to the detriment of the value of the REP token itself. Whether or not this is a structure that the broader community will find attractive over the long term will require more time to fully understand. HODLing Governance Augur’s governance structure will be familiar to anyone with knowledge of . A primary risk is obviously a series of disputes relating to a particular market that escalates to the point of forcing a . The team has made clear multiple times that the fork state possibility (laid out in the r) is encoded into the protocol simply to ensure that consensus will always be achieved, even though they ascribe a low likelihood to its ever being used in practice given the threshold required to prompt the fork. REP also suffers from risk exposure related to being built atop Ethereum. As Ethereum is in the midst of a significant, multi-year upgrade, any issues that may arise during the transition could also affect the Augur protocol. Ethereum fork whitepape Vulnerabilities Augur’s open-source development infrastructure is commendable, replete with easy to navigate GitHub pages and well-written documentation. While the worry of a 51% attack on REP is not irrational given the size of REP’s market, the incentives are not aligned to make such an attack likely. These facts paired with Augur’s solid history of security audits and active bug bounty program make the protocol attractive from a trust perspective. Augur’s biggest vulnerability may, in fact, be and , not a malicious attack. Without finding any product-market fit, the project is susceptible to new competition or the cryptocurrency space passing them by as new use cases for the technology are tested and implemented. UI issues lack of adoption Network effects While work on the Augur project has been in flight for years, it bears remembering that the mainnet itself has only been in operation since 2018. Given how nascent that makes the application, its growth in v1 remains fairly impressive and its impact as the first decentralized prediction market to successfully launch is groundbreaking. However, the community surrounding Augur realized improvements were needed in order to truly become successful and gain meaningful adoption. Because of this, the team took on the challenge of overhauling the current protocol in favor of a v2 to launch in 2020. It is still very early in the life-cycle of decentralized prediction markets, and more time is needed to see if Augur can grow its current users and scale sustainably in usage. Finally, it bears remembering that while usage of the dApp is important, REP’s popularity and listings on prominent exchanges are also key indicators. That said, given the limited utility of REP (market resolution within Augur), it is difficult to envision REP producing wide-ranging network effects in the near future. Team The three co-founders of Augur are , , and . While the founders at first glance may appear young and mostly inexperienced, they have behind them perhaps the most impressive advisors that a decentralized prediction market could hope for in , , and . The development team is diverse, approaching 30 members with ten different developers providing over the years. The project has maintained high levels of activity since 2017 projecting signs of a healthy, motivated project. Nonetheless, the lack of clarity surrounding who precisely is a part of the paired with the departure of certain key staff like Jeremy Gardner after Augur’s ICO cannot go unnoticed. Jeremy Gardner Jack Peterson Joey Krug Vitalik Buterin Elizabeth Stark Ron Bernstein significant commits Forecast Foundation OU, User experience Augur is and supports a variety of wallet solutions including MetaMask, , , AirBitz, and uPort, making REP no more difficult to secure from theft than any ERC-20 token. The wallet of choice when using Augur presently appears to be MetaMask given the extra control it grants users when it comes to gas price selection. Additionally, users can easily reference the and to track transactions or conduct their own audits. non-custodial Ledger Trezor source code block explorer As admitted by the co-founder of Augur himself, the protocol is currently hamstrung by one main user experience flaw: the various fees involved with participating in the decentralized prediction market. The dApp itself is beautifully designed, quick, and performs reliably for the most part, but fees do need to come down in the long run for market participants to migrate to Augur as their go-to prediction market. Before the self-imposed cutoff in , Augur v1 facilitated thousands of finalized markets with even more pending until v2 launches demonstrating that despite modest overall numbers, continued interest remains from some users. As A looks set to rollout sometime in 2020, the upgraded protocol addresses inherent in the original version that include reducing payout times for users and the currency risk (volatility) associated with ETH as the underlying asset to make wagers. Q3 2019 ugur v2 several UX weaknesses Regulation Perhaps the most interesting thing to note when it comes to regulatory risk to Augur’s future is the overall state of the regulation of online betting in the U.S. While online prediction markets are still in many respects a legal no-man’s-land, there are for proponents of online prediction markets and gambling. Augur’s management and appear unconcerned by potential regulatory risks facing the protocol. That said, the ambiguity of present applicable law and potential for downside risk remain concerns to be aware of. encouraging signs counsel Another regulatory hurdle for the REP token could be the question of whether or not REP can be classified as a security. This is one of the facing most crypto assets today and given that Augur raised money and distributed REP via an ICO only works against it in this case. It is difficult to assess the likelihood that any cryptoasset like REP would be classified as a security given the paucity of precedent, but the innovations behind Augur do not make its classification as a security most polemical risks any less likely Roadmap Unlike many projects in the crypto space that feature their full roadmap with anticipated dates prominently on all of their material, Augur version 1 opted instead for a system of distributed via their official Medium page until October 2019. Once the decision was made to overhaul the project in late 2018 and launch a , the latest updates for the transition can be found on their . From such posts, like this one outlining the view for , we can glean that some long-term goals of Augur include mobile UI improvements, integration, state channel trading, utilizing decentralized oracle Uniswap, off-chain trading via the , and improvements to the dispute resolution process. Substantial progress has been made towards v2, like the , and the official date for the launch of v2 has been slated for early weekly reports v2 website Augur v2 DAI 0x protocol cutoff of v1 June 2020. The lack of a consistently-running roadmap in a single location is a slight annoyance when it comes to attempting to track Augur’s future progression and v2 progress. This makes it difficult to know precisely when major developments might reasonably be expected as is the current case with v2. Originally planned to be released in late 2019 and then again in Q1 2020, Augur v2 has yet to fully release to the public.

AIRFIT

airfit.co

-- employees

Since n.d.

COMPANY RANKING

AIRFIT

EVERGREEN INDEX #655

Representing this company? Click here to claim and customize this page!

AirFit WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!

AIRFIT

airfit.co

-- employees

Since n.d.

COMPANY RANKING

AIRFIT

EVERGREEN INDEX #655

RELATED COMPANIESRANK

STORIES ABOUT AIRFIT

Representing this company? Click here to claim and customize this page!

AirFit WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!