Connect with us

Technology

How feature stores can reduce the ‘Groundhog Day’ effect for data scientists

Voice Of EU

Published

on

Splice Machine’s Monte Zweben explains how feature stores can help cut down the monotonous parts of a data scientist’s job.

Many people pursue a career in data science because they love solving problems. But something called the Groundhog Day effect can limit that, according to Monte Zweben, CEO of real-time AI company Splice Machine.

Zweben, who previously worked as the deputy chief of AI at NASA’s Ames Research Center and sits on the advisory board for Carnegie Mellon University’s School of Computer Science, believes feature stores can help.

Click here to check out the top sci-tech employers hiring right now.

‘Spending all your time on monotonous work can lead to unhappiness with the job’
– MONTE ZWEBEN

Can you explain what the Groundhog Day effect is for data scientists?

Work as a data scientist follows a cycle: log in, clean data, define features, test and build a model. Except not all parts of the cycle are created equal; data preparation takes 80pc of any given data scientist’s time.

No matter what project you’re working on, most days you’re cleaning data and converting raw data into features that machine learning models can understand. The monotonous void of data prep blends hours together and makes each day identical to the one before it.

With one person, it’s annoying to have to repeat the same work all the time; with a team, each person building features slightly differently can lead to inconsistent results.

Does this effect pose an issue?

From a productivity perspective, it’s incredibly inefficient for one person to repeat their own work multiple times; that’s time and money spent on unnecessary tasks, which makes models slower to get up and running.

From an employee perspective, spending all your time on monotonous work can lead to unhappiness with the job and increase employee turnover. For the business as a whole, lacking a centralised data process can also lead to inconsistencies in business.

If different people are defining features differently across a company, this can cause models and business decisions to differ based on feature definitions. Lifetime value of a customer (LVC) is a great example. One team might define the lifetime value as a customer’s total past spending, while another might include the customer’s projected value in the LVC.

Inconsistent definitions can lead to preferential treatment in a company and affect customer retention in the long term.

What are feature stores? How can they benefit data workers?

A feature store is a shareable repository of features made to automate the input, tracking and governance of data into machine learning models. Feature stores compute and store features, enabling them to be registered, discovered, used and shared across a company.

A feature store makes sure features are always up to date for predictions and maintains the history of each feature’s values in a consistent manner, so that models can be easily trained and re-trained.

Feature stores enable total model transparency, guarantee consistent training and can serve models real-time updates of aggregate data sets.

How do feature stores work?

A feature store is a repository of features, feature sets, and feature values, along with their feature history. The feature store has a set of services that interact with this repository, which includes defining features, searching for features, retrieving the current value of features, associating meta-data with those features, defining a training set from groups of features, and backfilling new features into training sets.

In some implementations, feature stores have user interfaces that call those services, and in others they are just APIs.

Feature stores are fed by pipelines that transform raw data into features. These features can then be defined, declared into groups, and assigned meta-data that makes them easier to search for. Once the features are in the store, they are used to create training views, training sets and serve features. These mechanisms allow feature stores to automate data transformation, serve aggregate features in real time and monitor models in real time.

How would you recommend data workers get on board with feature stores? What tips would you give them?

My number-one recommendation is to prepare for the future. Even if you only have a few models in production right now, I’ve seen so many data workers struggle to scale an ad-hoc data architecture. Within 10 years, the most successful companies will have hundreds and thousands of machine learning models running simultaneously; this will be impossible to manage without a feature store.

If you’re on the fence, just try one out! They’re easy to use and will seriously change your data workflow in the best way possible. I personally recommend trying a SQL-powered feature store. Make sure you get a feature store that works on any cloud and on premise, too. You don’t want to get locked into a specific cloud that might be less competitive in the future.

Are there any resources on the topic you would recommend?

I’m actually writing a book on feature stores for machine learning with Manning Publishers (scheduled release in 2021).

Featurestore.org is a great central location for lots of information about feature stores. The Toward Data Science blog on Medium has some great content on feature stores, too.

Source link

Technology

Chinese developers rebel against 996 working hours culture • The Register

Voice Of EU

Published

on

Chinese software developers have crowdsourced a spreadsheet that dishes the dirt on working conditions at hundreds of employers.

Dubbed WorkingTime, the protest aims to offer transparency regarding how many work hours are expected. Many organisations expect 72-hour working weeks – an arrangement dubbed “996” after the 9am to 9pm, six days a week culture in place at many Chinese companies.

The practice has sometimes been promoted by the rich and famous: Alibaba’s Jack Ma publicly stated that employees should actually want to work long hours and a job you love enough to spend that much time doing is a “blessing”.

Chinese courts take a different view. A recent decision found 996-style hours aren’t permissible, as Chinese law caps overtime at 36 hours per month and requires compensation for the extra hours. But China is not a workers’ paradise, and the practice persists because oversight is limited and independent labour unions are illegal in the Middle Kingdom.

The WorkingTime project aims to assist developers looking for work to understand what they’re signing up for.

“The opacity of working hours in some companies, working time is a very important factor in choosing an offer,” wrote a movement founder on Chinese Q&A site Zhihu.

The spreadsheet in which workers record how many hours they work a week, job descriptions, breaks and other remarks strongly suggest that grueling hours remain at some workplaces. Others stick to a 40-ish hour working week and add perks like happy hours and subsidized housing.

The anecdotes, visible on an openly accessible spreadsheet associated with the project, provide a similar service for Chinese tech workers to web pages like Glassdoor – giving tips on company culture and requirements.

Some remarks include:

  • “I often go on business trips. I have been on business for half of a month. I leave work after 10 o’clock every night at the customer’s site. I have to work overtime on weekends. The entire department has worked for two years except for the leaders.”
  • “Feel free to ask for leave and lunch time, because it’s the field work, whether you are in the company or not, and you can play games casually, regardless of the leader. If you drink too much, it’s fine if you don’t come.”
  • “Mandatory to keep people on duty every night, compulsory all staff to work overtime every Saturday, no overtime pay, working hours over 10 hours.”
  • “When the daily work cannot be completed, it is necessary to work overtime at home.”
  • “The work pace is fast and the work content is highly saturated. Flexible commute, just do everything.”

The WorkingTime project has gone viral, with the founders reporting over ten million views and thousands of entries as of last Tuesday. While the founders remain anonymous, contributors hail from a diverse subset of companies that includes some of China’s big tech giants like Alibaba, Tencent, Huawei and Bytedance, as well as multinational companies such as SAP, Dyson, Intel and IBM.

According to the project’s GitHub page, lawyers are currently pitching in to sort out legal issues prior to making the project freely downloadable. However, a summary table of data collated daily is already available in Chinese.

Unsurprisingly, the project has stirred some ire. The founders have asked that participants do not apply for editing permission, explaining that “due to malicious editing” such privileges will not be granted. ®

Source link

Continue Reading

Technology

Johnson & Johnson Ireland moves to 100pc renewable electricity

Voice Of EU

Published

on

The move comes following a power purchase agreement between Johnson & Johnson and Ørsted, which has windfarm sites in Clare and Kerry.

Johnson & Johnson has revealed plans to move to 100pc renewable electricity across its Irish operations.

The company has entered into an eight-year corporate power purchase agreement in Ireland with Danish company Ørsted. The agreement will help to ensure that the company’s entire Irish operations will be powered by electricity from 100pc renewable sources from now on.

Ørsted will supply the company with more than 1TWh of renewable energy during this period from two windfarms located in Kerry and Clare. The agreement will also help Ørsted as it invests in its strategy to construct more renewable generation in the future.

Taoiseach Micheál Martin, TD, praised the move in the context of Ireland’s climate action plans.

“Johnson & Johnson has embraced its environmental responsibilities globally, but also here in Ireland, and this agreement will help the company to achieve its wider climate goals. We are at a crucial point in the global fight against climate change and initiatives like this should become the benchmark for all companies to aspire to,” he said.

Towards net zero

Last year, Johnson & Johnson’s worldwide VP of environmental health, safety and sustainability, Paulette Frank, spoke at Silicon Republic’s Future Human event about the company’s “bold” climate goals. From her base in the US, Frank told attendees of the virtual event that her colleagues viewed the pandemic as “inspiration to propel” its climate action “further faster.”

Sourcing electricity from 100pc renewable sources is a goal the company set to achieve by 2025. By 2030, it wants to achieve carbon neutrality in its global operations.

John Lynch, plant leader at Johnson & Johnson Vision Care Ireland, said the company was proud to have met its targets in its Irish operations.

“Across our 10 sites and workforce of more than 5,000 here in Ireland, we are committed to supporting Johnson & Johnson’s climate action goals. In the last decade we have invested more than €60m in over 80 carbon footprint reduction projects.

“Today is a major landmark on our journey in Ireland to achieve net zero carbon emissions by 2030 and underlines our commitment to ensuring a better, healthier world.”

Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.

Source link

Continue Reading

Technology

‘At once intolerable and addictive’: five wellbeing courses and apps, road-tested | Health & wellbeing

Voice Of EU

Published

on

Australians are the world’s biggest consumers of health and wellness apps, punching well above our per capita weight in our quest for peak physical and mental condition, according to research from telecommunications company Uswitch. In recent years we have also been making them – with everyone from fitness influencers to mental health advocacy groups launching digital products.

I’m partial to a bit of mobile-based movement and mindfulness myself, but I have a complex relationship with wellness. While I love green juices, pilates and my “ness” being “well”, I can’t abide many contemporary uses of the word. In the diet, fitness, fashion and other industries, “wellness” can feel like a barely repackaged “weight loss”, while “healthy” has replaced “slim” as companies respond superficially to the body positivity movement without really changing their ways.

Despite wholesome beginnings in the 1950s, wellness is often framed as a goal for the financially and genetically privileged – and don’t get me started on the pseudoscience.

So I choose cautious cynicism when engaging with wellness and wellbeing products – but I’ve also been alone in my house for the greater part of two years, so I’ll try pretty much anything.

Sweat

Cost: $19.99 a month

Screen shot of the Sweat app from Kayla Itsines.

Sweat is a women’s health app co-founded by Australian fitness influencer Kayla Itsines, who boasts a worldwide social media following of more than 40 million. It offers over 30 programs for training at home or the gym, including high-intensity interval training (Hiit), low-intensity training, yoga and barre.

I did sessions from the PWR Zero Equipment program and it was all easy to follow and very doable. Audio and written instructions and onscreen demonstrations are clear, and self-accountability is super easy. It’s perfect for lockdown and for busy people cramming in exercise wherever and whenever they can. Plus, I can report that burpees are still the merciless work of Satan herself.

Itsines has created an app that exists in the wellness space with little of the self-congratulatory, quasi-spiritual hoopla other influencers lean so heavily into. Sweat isn’t pretending to be something it’s not. It’s a workout app, you do workouts on it. Yes, there are recipes and lifestyle tips but they aren’t offered as miracle pathways to a higher plane of being.

Is it my preferred mode of exercise? No. But it’s convenient and flexible and I can see myself using it when I travel. If that’s a thing that ever happens again.

Worry Time

Cost: Free

ReachOut’s WorryTime app
ReachOut’s WorryTime app. Photograph: Reach Out

ReachOut’s WorryTime is an anxiety management app from the online youth mental health service that uses cognitive behavioural therapy (CBT) techniques to disrupt and manage repetitive thinking.

I am by no definition a youth, but I have mild anxiety and WorryTime’s methodology appealed to me. You nominate a daily time to do all your worrying and when you feel anxious, you note why in the app; every day at the designated time, you worry about what’s still plaguing you and delete what’s not. Easy!

I used WorryTime diligently for a while, noting my fears, my troubles and doubts and reassessing them every 24 hours. All was going well until I got busy with work, stressed about work and scared I’d stop getting work. Where the app had been a welcome task, it became a bugbear.

I was trying not to think about things that made me anxious and knowing the app contained a list of them created a classic avoidance paradigm. I skipped a day. And the next day. And the day after that. Soon the WorryTime alarm was causing me the very anxiety it was engaged to minimise. After a few weeks of this mental chicken-egg dance, I deleted the app. I may have been in the foetal position at the time.

I’m not advocating against WorryTime. It could be a great tool for others. There are no one-size-fits-all mental health salves. It would be nice if there were though.

Bibliotherapy with State Library Victoria

Cost: Free

Dr Susan McLaine, host of State Library Victoria’s Bibliotherapy podcast
Dr Susan McLaine, host of State Library Victoria’s Bibliotherapy podcast. Photograph: Supplied

My favourite discovery from this whole exercise is bibliotherapy or book therapy, an age-old practice that uses literature to support better mental health and wellbeing. Basically, you read or are read aloud a prescribed text, specifically chosen to raise questions, uncover truths and encourage healing. It’s also fun to say.

In response to the pandemic, a new podcast called Bibliotherapy with State Library Victoria was launched. Hosted by bibliotherapy practitioner Dr Susan McLaine, it offers to help people “stay calmer in this fragile time”. In each episode, McLaine reads a short story and a poem and poses questions for listeners. Texts range from emerging and obscure writers to Tolstoy, Donne and Kipling.

I love this podcast. There’s something so intimate and soothing about being read to, no doubt embedded in childhood nostalgia. McLaine’s voice takes some getting used to, though to be fair I find this with most podcast hosts, but her choice of texts is excellent and she reads everything slowly and deliberately, “savouring every word and offering space between words”. It’s the closest thing to a hug I’ve had in months.

The only bad thing about it is that there are only two short seasons. After a brief search for similarly soporific, story-based podcasts and apps, I found the excellent Dreamy podcast, a collection of beautiful sleep stories by First Nations storytellers like Jazz Money and Aurora Liddle-Christie. Bringing tens of thousands of years of oral tradition into the digital world, Dreamy is “helping people of all walks of life to quiet their minds, drift into dreams, and disconnect from their devices”.

I also found Sleep Stories on the Calm app ($14.99 a month). It’s full of grown-up tales and mindful nonsense to soothe or bore you into slumber. There are even equally terrible and amazing celebrity cameos: Matthew McConaughey, Cillian Murphy and the hot duke from Bridgerton will read to you like you’re a child. Last night Harry Styles read me the worst poem I’ve ever heard – for 40 minutes. Five stars. Would listen again.

The Resilience Project

Price: $4.49 one time fee

The Resilience Project Wellbeing App.
Photograph: Supplied

The Resilience Project app is a “daily wellbeing journal” for all ages from a Melbourne-based organisation of the same name, providing evidence-based mental health strategies and “sharing the benefits of gratitude, empathy and mindfulness” to schools, sports clubs and businesses.

Users are encouraged to log on every day, note how they feel, record who or what they’re grateful for, perform acts of kindness and do a short guided meditation. This nice daily ritual only takes a few minutes but proves a small antidote to the current news cycle.

I don’t see myself using it long-term, because of repetitiveness and the world’s shortest attention span, but during this lockdown I’ve appreciated the nightly reminder to acknowledge my blessings and privilege and to reach out to friends.

Though it can’t do the heavy lifting where mental health is concerned, I’ll put it in my arsenal of chronic depression coping mechanisms, and try to use it in bad times. It won’t soothe what only drugs and Great British Bake Off can, but it might provide a few minutes respite.

The Class

Cost: $40 a month

The Class Digital Studio is a mat-based exercise program, with elements of yoga, pilates, cardio, free-style dance, expansion, and release.
The Class Digital Studio is a mat-based exercise program, with elements of yoga, pilates, cardio, free-style dance, expansion, and release. Photograph: The Class Digital Studio

The Class is an American exercise methodology-slash-mindfulness practice with semi-cult vibes, taught by a host of ridiculously hot and relentlessly cool twentysomethings who can pull off white Lycra and blend in on a Girls set.

In fortuitous timing, founder Taryn Toomey launched online classes in late 2019, taking the Class into locked down homes around the world from 2020. Australians can access a wide selection of on-demand and live online classes, and there’s even an Australian teacher. Timezone differences narrow live options quite a bit, but most live classes become on-demand classes, so it doesn’t really matter.

Frequented by celebrities including Alicia Keys, Naomi Watts and Emma Stone, the Class is a mat-based, music-driven “cathartic workout experience” designed to “strengthen the body and balance the mind”. It’s yoga meets Les Mills meets clubbing. Movements are simple, repetition is key and loud exhales are encouraged. You may do squats for a whole song, free dance for another and star jumps for the next. In between, there’s stillness.

Teachers speak a kind of motivational psychobabble that is at once intolerable and addictive. It verges on the spiritual and flirts with cultural appropriation but remains just secular enough that I don’t turn it off. “Be in your power”; “You are enough”; “Softness is your birthright” and so on. Many teachers end their sessions with “I love you” which I somehow don’t hate.

At first, I struggled to put aside my prejudices against self-indulgent, pseudo-mystical wellness fads and find peace with beautiful women telling me to accept myself while making me do burpees. But the more I did it, the more I was able to just let go and roll with the theatre. Plus, it’s actually a very good workout.

I am now willingly paying for the Class. Let’s never speak of this again. I love you.

Source link

Continue Reading

Trending

Subscribe To Our Newsletter

Join our mailing list to receive the latest news and updates 
directly on your inbox.

You have Successfully Subscribed!