Show HN: Augmenting tabular data with SDV to improve ML results https://bit.ly/2Zdbp4w

Show HN: Augmenting tabular data with SDV to improve ML results https://bit.ly/3rU8GcN February 12, 2021 at 06:51PM

Show HN: A Python Libary That Simplifies Data Validation https://bit.ly/2NoTl4F

Show HN: A Python Libary That Simplifies Data Validation https://bit.ly/3tT3TKk February 12, 2021 at 05:21PM

Show HN: Marketing Patterns – DIY Template for Growth https://bit.ly/3aYQGqF

Show HN: Marketing Patterns – DIY Template for Growth https://bit.ly/3qi1QNx February 12, 2021 at 06:42PM

Show HN: Minimalist Tor-to-Web Proxy https://bit.ly/3af71bJ

Show HN: Minimalist Tor-to-Web Proxy https://bit.ly/3acjahx February 12, 2021 at 05:52PM

Show HN: DeFiOptions – ERC20, collateralized, cash settable options on ethereum https://bit.ly/2MYn98B

Show HN: DeFiOptions – ERC20, collateralized, cash settable options on ethereum https://bit.ly/2ZbNJO4 February 12, 2021 at 12:50PM

Show HN: Digital Pottery in the Browser https://bit.ly/3rNErny

Show HN: Digital Pottery in the Browser https://bit.ly/3rKspLM February 12, 2021 at 01:05PM

Show HN: Create animated and interactive drawings in the browser https://bit.ly/3pepkSz

Show HN: Create animated and interactive drawings in the browser https://bit.ly/3tRHcWV February 12, 2021 at 01:04PM

Show HN: Zig Xor Filters (“Faster and Smaller Than Bloom Filters”) https://bit.ly/3b33Kv6

Show HN: Zig Xor Filters (“Faster and Smaller Than Bloom Filters”) https://bit.ly/2NnOuRf February 12, 2021 at 09:55AM

Show HN: I built an URL shortener to share Clubhouse events https://bit.ly/3pdrAcS

Show HN: I built an URL shortener to share Clubhouse events https://bit.ly/3qe8Xa3 February 10, 2021 at 05:34PM

Show HN: Easy Dependency Injection for Golang https://bit.ly/3tQz5tO

Show HN: Easy Dependency Injection for Golang https://bit.ly/3q8xf57 February 12, 2021 at 01:57AM

Show HN: Real-time multiplayer games with cubes. Early feedback on dev docs? https://bit.ly/3rP8Wd1

Show HN: Real-time multiplayer games with cubes. Early feedback on dev docs? https://bit.ly/3tOghLF February 12, 2021 at 12:51AM

Show HN: EffectNode Studio, trying to make complex graphics code more manageable https://bit.ly/375nEES

Show HN: EffectNode Studio, trying to make complex graphics code more manageable https://bit.ly/3addjZu February 11, 2021 at 11:57PM

Show HN: git-peek – git repo to local editor instantly https://bit.ly/3phpbgZ

Show HN: git-peek – git repo to local editor instantly https://bit.ly/2OuR4pn February 11, 2021 at 11:07PM

Show HN: Sleepy Time Conference—conferences that comes together while you sleep https://bit.ly/373ByHe

Show HN: Sleepy Time Conference—conferences that comes together while you sleep https://bit.ly/3pauYVB February 11, 2021 at 10:50PM

Show HN: Nirvana.Work – Automated Task Scheduling – Put Sprints on Autopilot https://bit.ly/3qbUawx

Show HN: Nirvana.Work – Automated Task Scheduling – Put Sprints on Autopilot https://bit.ly/3qdalJU February 11, 2021 at 06:08PM

Launch HN: Cord (YC W21) – training data toolbox for computer vision https://bit.ly/2Z7vuJB

Launch HN: Cord (YC W21) – training data toolbox for computer vision Hey HN community - I’m Ulrik from Cord ( https://bit.ly/3qdaflw ) in the current YC W21 batch [1] - we are building software that allows people to label their data intelligently using a toolbox of various ‘labeling algorithms’. Labeling algorithms are any units of intelligence (e.g. a pre-trained model, or an interpolation algorithm) that help automate the annotation process. This enables data science and machine learning teams to rapidly iterate on their ML models without having to farm out labeling tasks to an external workforce. Today we’re launching the first part of our product, our Web App, which serves our initial set of automation features through a GUI. It also allows you to classify images and draw vector labels, visualize data, and perform collaborative QA. Computer vision ML algorithms are widely used for tasks like detecting everyday objects such as cars and pedestrians. However, they are yet to see widespread adoption for things like detecting cancerous polyps during an endoscopic procedure or blood clots in MRI scans. The lack of massive-scale labeled training datasets that fuel contemporary approaches is often the blocking element in building ML applications that solve these more specialised tasks. We also believe that the core part of the IP of an ML application stems from the labeled data used to train it. Creating these datasets is challenging for several reasons. Labeling the data requires expensive domain-expert annotators, and privacy might prevent the data from being sent to an external workforce. Ultimately most labeling work tends to be done using open-source tools that were not created for speed and purpose-built to handle massive-scale datasets[2]. These tools also tend to provide a poor experience for the end consumer of the training data (e.g., data scientists, ML engineers) because they lack intelligence and require high manual input. The initial seed of the idea came while I was working on a CS master’s project of visualizing massive-scale medical image datasets. I saw saw how much time and effort was being spent by doctors on labeling data. I met my co-founder Eric, who had worked as a quant researcher in finance, and after meeting him we realized we could take an algorithmic approach to tackling the labeling problem. Instead of writing trading algorithms, we turned our focus to writing labeling algorithms. For example, for a food calorie estimation project we translated image level classifications of food items to individualized bounding box labels using a labeling algorithm we wrote with our SDK, requiring only one manual label per food item. Although it was an image dataset, our algorithm approximated noisy bounding box labels by using a CSRT object tracker across images. It then trained a shallow Faster RCNN ‘micro-model’ on the noisy labels, ran inference on the data, and suppressed earlier noisy labels. We then quickly visually reviewed and adjusted the results on our Web App[3]. We have applied a similar approach in areas such as gastroenterology[4] and pathology. The days of relying on an army of human annotators and waiting to start the model building process are hopefully (soon) over. We are incredibly excited to be driving for that change - and are delighted to be sharing Cord with the HN community! We would love to hear your feedback. How are you going about creating and managing training data today? What are your key constraints? If you have used a creative method to label your data before, please share. Thank you so much in advance! [1] What I Learned From My First Month at Y Combinator - https://bit.ly/3a9Osp3... [2] Why You Should Ditch Your In-House Training Data Tools (And Avoid Building Your Own) - https://bit.ly/3d71DZU [3] Label a Dataset with a Few Lines of Code - https://bit.ly/3aV3w9h... [4] Pain Relief for Doctors Labelling Data - https://bit.ly/3b0cbHP... February 11, 2021 at 06:06PM

Launch HN: Chorus Meditation (YC W21) – Meditation for Non-Meditators https://bit.ly/3jDau73

Launch HN: Chorus Meditation (YC W21) – Meditation for Non-Meditators Hey everyone! I’m Ali and, together with my co-founders MK and Warren, I’m building Chorus Meditation ( https://bit.ly/3jDauUB ). We provide online group meditation classes led by trained instructors. MK and I met after we both had found the benefits of a traditional meditation practice, but only after much difficulty getting started because it took over 30 days to feel the benefits and it can often feel isolating and like nothing is “working.” At the time, MK was a top SoulCycle instructor. She is a true master at creating community and motivating people to be their best through a perfect balance of humor, approachability, and deep vulnerability and acceptance. I was an avid SoulCycle rider and we bonded over our shared love for meditation and separately, our love for the instantly gratifying and social experience that SoulCycle had created. She and I decided that if we could create an experience for the mind that mirrored what SoulCycle had done for the body, we could help millions of people just like us. So, we spent months, combining different mindfulness techniques into a new method, testing out various versions on our living floors. We tried starting the class with a 3 minute traditional meditation before moving into the breathing pattern - no dice - we had promised people non-traditional meditation so when we hit them with exactly traditional meditation right at the start, it turned people off. Next we tried getting into the breathing pattern right off the bat -- still no dice. But we kept at it, and 16 major iterations later, we landed on what is now our Chorus class. Traditional meditation can be life-changing for those who stick with it, but the unfortunate truth is that for most people it’s hard to sustain the discipline to stick with it long enough to unlock the ah-ha moment. Once you cross that threshold you feel its power, but with Chorus we are trying to help people who struggle with that onboarding phase cross the threshold more easily. We've found that one of the main barriers many people run into with traditional meditation is that they're doing it alone, and they often feel like nothing is happening. So, we made Chorus 1) social, with warm, personable teachers and fellow class attendees, 2) fun, with new and popular music, and 3) designed to give motivating results in the first session and on-going. For example, the breathing pattern we use brings more oxygen into the body than normal inhales and exhales, which causes a tingling sensation, giving users a quick and satisfying feeling even in the first session. You can think of the tingles like endorphins in exercise - they feel good and tell you that something is working - so you are satisfied and want to come back for more. Everything in Chorus is designed to motivate you to keep going. Our members pay $40-a-month to have access to live and pre-recorded classes set to the beat of popular music like Beyonce, Odesza, Bon Iver, etc, that help them start their day with a positive mindset or unwind at night before bed. If you want to give it a try, we just launched a new class specifically designed to help you sleep — https://bit.ly/3p96yMh One of our users, a mother of young twins, shared: “my first experience unlocked something in me. Something visceral, and I thought - ‘this is so worth exploring.’” This is exactly the kind of reaction we’re going for. I want to emphasize that we’re in no way trying to replace traditional meditation. We, ourselves, are reverent students of traditional practices. And we're well aware that we don't have anything to teach the millennia-old traditions of India and China. What we are trying to do is bridge the gap for people who find traditional techniques challenging so that they can avoid the discouraging feeling of “I’m doing this wrong” and empower them to develop their own mindfulness practice. Another thing we do to support our users in the early stages of practice is provide a community in which they can share their experiences and get encouragement to keep going. This is one of the more satisfying aspects for us, because people report their positive experiences as well as their challenges. We hear from users who report feeling more calm and focused, or sleeping better, all the way up to "Chorus has truly transformed my life...I didn’t think I would ever have a relationship with my mom again, and now because of Chorus, I do.” We are building Chorus for our collective community, so I’d really love to hear this community’s feedback. We'd love to hear from everybody, whether you're a complete meditation skeptic, someone who's found meditation challenging, or a seasoned meditator who has achieved total equanimity! We're eager to hear your experiences and thoughts and feedback! Over to you, HN! February 11, 2021 at 05:25PM

Show HN: GitHub Repo to Local Editor In Less Than 1 Second https://bit.ly/3rH9Zvo

Show HN: GitHub Repo to Local Editor In Less Than 1 Second https://bit.ly/2MMz9Kp February 11, 2021 at 05:24PM

Show HN: Spacelift – first all-in-one CI/CD for Infrastructure as Code https://bit.ly/2Nf5edI

Show HN: Spacelift – first all-in-one CI/CD for Infrastructure as Code Hi HN! We are the team behind Spacelift (https://bit.ly/3qerjHL). Spacelift is the CI/CD for infrastructure-as-code, be it Terraform, Pulumi, CloudFormation or Ansible (coming soon), and policy as code. It enables collaboration, automates manual work and compliance, and lets teams customize and automate their workflows. Here’s what you can do with Spacelift - Build sophisticated Git-based workflows - Use Open Policy Agent to declare rules around your infrastructure, access control, state changes, and more - Author and maintain reusable modules for your organization; we even have a full CI solution for modules to make sure they’re healthy - Declare who can log in (and under what circumstances) and what their level of access to each of the managed projects should be (SAML 2.0 SSO out of the box!) using login and access policies respectively - Use Spacelift’s trigger policies to create arbitrary workflows and dependencies spanning multiple infrastructure-as-code stacks - Manage stacks, contexts, modules, and policies in a declarative way using Terraform or Pulumi Before Spacelift, we built bespoke solutions (e.g., Geopoiesis, https://bit.ly/3rJvWdj), currently used by two of the largest European scaleups. In the past few months, we’ve been onboarding our first customers and making sure everything works as expected. You can check out our starter repo at https://bit.ly/2MWn5pR. It's an easy way to learn all of Spacelift’s capabilities in 15 minutes without tapping into your own cloud resources. We’d love your thoughts on our approach and anything that has worked or hasn’t worked for you. P.S. We are hiring https://bit.ly/3qerlzn P.P.S. We just announced our funding round https://tcrn.ch/3qdqEqb February 11, 2021 at 04:36PM

Show HN: Next.js Template for Interactive Courses https://bit.ly/3p5CcKE

Show HN: Next.js Template for Interactive Courses https://bit.ly/2LHluDR February 11, 2021 at 02:21PM