Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • 06:57
    guillaume-chevalier edited #319
  • 06:56
    guillaume-chevalier edited #319
  • 06:56
    guillaume-chevalier edited #319
  • Dec 01 18:47
    LuisBlanche starred Neuraxio/Neuraxle
  • Nov 30 19:56
    vincent-antaki commented #424
  • Nov 30 19:13
    cla-bot[bot] commented #424
  • Nov 30 19:13
    vincent-antaki synchronize #424
  • Nov 29 20:25
    cla-bot[bot] commented #424
  • Nov 29 20:25
    vincent-antaki synchronize #424
  • Nov 26 18:20
    vincent-antaki closed #405
  • Nov 26 18:20
    vincent-antaki commented #405
  • Nov 26 03:42
    vincent-antaki edited #424
  • Nov 26 03:40
    vincent-antaki edited #424
  • Nov 26 03:36
    cla-bot[bot] commented #424
  • Nov 26 03:36
    vincent-antaki synchronize #424
  • Nov 26 03:17
    cla-bot[bot] commented #424
  • Nov 26 03:17
    vincent-antaki synchronize #424
  • Nov 26 02:48
    cla-bot[bot] commented #424
  • Nov 26 02:48
    vincent-antaki synchronize #424
  • Nov 26 02:31
    cla-bot[bot] commented #424
Sameroom
@sameroom-bot
[IFTTT]
New release from git tag: 0.3.0
New release from git tag: 0.3.0
Sameroom
@sameroom-bot
[IFTTT]
Neuraxio's ML Blog
A Rant on Kaggle Competition Code (and Most Research Code) https://ift.tt/2ZwmUna
Sameroom
@sameroom-bot
[IFTTT]
Neuraxio's ML Blog
Why Deep Learning has a Bright Future https://ift.tt/2Q7Hbw6
Sameroom
@sameroom-bot
[IFTTT]
Neuraxio's ML Blog
What's wrong with Scikit-Learn. https://ift.tt/36ogN6O
Sameroom
@sameroom-bot
[IFTTT]
New release from git tag: 0.3.1
New release from git tag: 0.3.1
Sameroom
@sameroom-bot
[IFTTT]
New release from git tag: 0.3.2: Merge pull request #268 from guillaume-chevalier/master
New release from git tag: 0.3.2: Merge pull request #268 from guillaume-chevalier/master
Sameroom
@sameroom-bot
[IFTTT]
Neuraxio's ML Blog
Your Data Scientist is a Mom https://ift.tt/3bYuixb
Sameroom
@sameroom-bot
[IFTTT]
New release from git tag: 0.3.3
New release from git tag: 0.3.3
Sameroom
@sameroom-bot
[Guillaume Chevalier, Neuraxio Open-Source] :tada:
Sameroom
@sameroom-bot
[IFTTT]
New release from git tag: 0.3.4
New release from git tag: 0.3.4
Sameroom
@sameroom-bot

[Guillaume Chevalier, Neuraxio Open-Source] Hello @Everyone!

I'm repurposing this slack workspace to a wider range of possible collaborations:

  • Collaborate with us and other researchers on writing more time series processing papers, in the #research channel;
  • Do business with us and other companies for services and products related to time series processing, in the #business channel;
  • Talk about how to do Clean Machine Learning using Neuraxle, in the #neuraxle channel;

More to come!
:confetti_ball::tada:

Sameroom
@sameroom-bot
[Rob Romijnders, Neuraxio] Hi all! Thanks for inviting me to this channel. Except for Guillaume, I don't know anyone in this channel. What are your backgrounds and research interests?
Sameroom
@sameroom-bot

[Guillaume Chevalier, Neuraxio] Thank you @Rob Romijnders for starting the discussion! Here is an intro:

@Rob Romijnders works a lot with time series, RNNs and seq2seq. I wrote a paper with him in the past about predicting Gaussians as outputs of RNNs to predict basketball trajectories in 3D.
@Moloud Abdar wants to predict confidence intervals and apply to big conferences to then launch a startup as he plans. He also collaborates with other researchers and interns.
@Eric Hamel is coding the TPE algorithm in Neuraxle and he likes Automated Machine Learning (AutoML), I've collaborated with him in the past as well
@Alexandre Brillant is working with me on Neuraxle and on other business R&D projects involving RNNs, we've known each other for 10 years approximately
@ is working on Spiking Neural Networks, and to complete his Ph.D. he'd like to very soone write a paper for the mathematical proof to complete what I've advanced here: https://github.com/guillaume-chevalier/Spiking-Neural-Network-SNN-with-PyTorch-where-Backpropagation-engenders-STDP - it is to be noted that this article of mine don't yet have a corresponding paper, and it'd be a good opportunity to finally write a paper about this work I did. Mickael already have improved the hyperparameters, and he'd help a lot in doing the actual writing and publishing.
@Yassine and @Ahmed Abassi created the NEURODATA startup in Tunisia, they'd like to collaborate with me on some research projects, they suggested Kaggle competitions as a starting point.
@JeromeBlanchet loves open-source projects, and we've almost co-founded Neuraxio as well, although distance was a problem at that moment
@ worked for me in a final school project using BERT, and in another business project using seq2seq neural networks.

@Everyone For the other ones whom I didn't introduce, it might be because I don't know you enough. Don't hesitate introduce yourself to us and to tell us what you'd like to do :)

I've also invited more people to this slack recently. I think we would have ways to collaborate on more papers and projects. Note that I see this slack as a good space to communicate and collaborate. This said, don't hesitate to invite anyone who would be interested in working on those projects with us. After discussing more, I'd see people move to private channels or something to collaborate more closely on projects and to avoid spamming everyone here too much.

Sameroom
@sameroom-bot
[Reda EL Hail, Neuraxio] Hello everyone, I am Reda from Morocco, I am an intern at InterDigital R&D in France. I am working on a project about estimating human activity based on Radar's data and I would like to apply RNNs for this purpose.
Sameroom
@sameroom-bot
[Delamare Mickael, Neuraxio] Hello everyone, I am Mickaël from France. I am engineer on mechatronics. I am currently PhD student at Siatech and IRSEEM laboratory i’m working on gesture recognition really interesting about SNN/RNN for this purpose. Looking forward to work together and have idea (bicep emoji).
Sameroom
@sameroom-bot
[Moloud Abdar, Neuraxio] Hi all, my name is Moloud Abdar, a Ph.D. student at Institute for Intelligent Systems Research and Innovation (IISRI), Deakin University, Australia. I am working on deep learning and machine learning. I would solve some important real issues with deep learning methods and submit results to very important conferences such as CVPR, NIPS, ICML, ECCV, ICCV, AAAI, ICLR, etc.
Sameroom
@sameroom-bot
[Yassine, Neuraxio] Hi everyone , Yassine Hamdaoui From Tunisia
I'm here to collaborate with @ in some research projects i had experienced many AI fields .. right now i'm working with RL . Tensorflow is my favourite framework .
I suggest also to start with kaggle competitions
We can team up and get engaged with the commy
Sameroom
@sameroom-bot
[Rob Romijnders, Neuraxio] Awesome. I currently work at Google Brain. My research focuses on representation learning from videos. That said, my personal heart lies with Bayesian deep learning. Anyone who want to collaborate on these topics, let's have a chat!
Sameroom
@sameroom-bot
[Imane, Neuraxio] Hello everyone, my name is Imane Lamiche from Algeria, a Ph.D. student at Northwestern Polytechnical University, China. I am working on behavioral biometrics for smartphone authentication, My main focus is to develop efficient time-energy consumption solutions. I would solve some important real issues with deep learning methods and submit results to very important conferences such as CVPR, NIPS, ICML, ECCV, ICCV, AAAI, ICLR, etc.
Sameroom
@sameroom-bot
[Guillaume Chevalier, Neuraxio] Nice to see we're already 25 in this slack! I've created some channels to get more organized. I've quickly invited some people to the channel yet from what I could observe from your introductions, but I haven't invited everyone, so please join the channels that interest you:
• #snn to work on Spiking Neural Networks. I invited @ since he has already started to improve my work. We'd like to do the mathematical proof of my claim that Backprop in SNN engenders STDP with the help of a mathematician, and he could write and publish the paper with his lab or startup.
• #activity-recognition for everything related to LSTMs and Activity Recognition. I've invited @Imane, @, and @Reda EL Hail as you seem directly interested in this topic.
• #bayesian-dl for Bayesian Deep Learning. @Moloud Abdar, @Rob Romijnders and @Yassine you seem to be quite interested in this topic. Let's spark up a discussion in this channel to see how we could collaborate.
[Guillaume Chevalier, Neuraxio] For the others, I'd be interested in knowing what you'd like to do and if you'd join one of those channels (or create new ones!). @Ahmed Abassi @Ayara @Cavan @ @DJ @Eric Hamel @Gude kiran @ @JeromeBlanchet @Karthik @Maxime Leclerc @Nathan Cheever
Sameroom
@sameroom-bot
[Ibrahim Ali Metin, Neuraxio] Hello friends
[Ibrahim Ali Metin, Neuraxio] my profile https://www.linkedin.com/in/ibrahim-ali-metin-5b55a3b3/
[Ibrahim Ali Metin, Neuraxio] I'm very glad to meet you
Sameroom
@sameroom-bot
[Delamare Mickael, Neuraxio] Hello ! :)
Sameroom
@sameroom-bot
[Zishi Wu, Neuraxio] Hello! Hope everyone is safe and healthy in these times. My name is Zishi Wu, I'm a PhD student in Computer Science at the University of Miami in Florida. This summer I am doing an internship with an insurance company and my team is trying to use machine learning models to forecast the quantity and cost of insurance claims. Last year I took a linear regression class where the book examples had pre-curated datasets containing 50 data points each and the prediction errors were normally distributed. This new task I am faced with is way more difficult because the data is not normally distributed and contains many extreme values. The regression models I've tried thus far (ridge regression and gradient boost) have consistently outputted predictions that underestimate or overestimate the extreme values. Linear regression models and their variations, which assume the errors to be independent and normally distributed, are failing big time. I think either a finely tuned XGBoost or neural network model should do, but I am looking for a robust pipeline for finding the optimal hyper-parameters of such models. I stumbled upon Neuroaxio when I was reading about sklearn pipelines on Stackoverflow and found an article "What's Wrong with Sklearn Pipelines". I'm very intrigued by this Python package and want to try it out for my project. However, I have no previous experience deploying production level machine learning models. There seems to be a lot of companies who are pitched solutions to quickly create and deploy machine learning models. Can someone explain to a beginner like myself: what's the difference between Neuroaxio and other companies products such as H2O.ai, fast.ai, Databricks AutoML, DataRobot, and a bunch of other companies whose names I just haven't heard of yet?
Sameroom
@sameroom-bot
[IFTTT]
New release from git tag: 0.4.0: Merge pull request #308 from alexbrillant/issue-236
New release from git tag: 0.4.0: Merge pull request #308 from alexbrillant/issue-236
Sameroom
@sameroom-bot
[Felix Levesque, Neuraxio] Whats up people! Data Scientist consultant from Montreal working for small companies on my own.
[Felix Levesque, Neuraxio] Trying to learn as much as possible from you guys and your research. Currently using seq2seq to predict time series for a consulting gig.
Sameroom
@sameroom-bot
[Ed, Neuraxio] from another quant channel https://www.youtube.com/watch?v=zQ_rL6DjWEU
Sameroom
@sameroom-bot
[IFTTT]
Neuraxio's ML Blog
Clean Machine Learning Training https://ift.tt/2Bwpmlm
Sameroom
@sameroom-bot
[Queen Saikia, Neuraxio] can anyone please tell me how to run this code https://github.com/guillaume-chevalier/HAR-stacked-residual-bidir-LSTMs
Also please suggest me the research paper for this one https://github.com/guillaume-chevalier/LSTM-Human-Activity-Recognition please its very important.
Sameroom
@sameroom-bot
[Shreshth, Neuraxio] hi all, I'm looking into HAR using 2d Poses. Any links to previous work or research papers are appreciated
Sameroom
@sameroom-bot
[IFTTT]
New release from git tag: 0.4.1
New release from git tag: 0.4.1
[IFTTT]
New release from git tag: 0.5.0
New release from git tag: 0.5.0
Sameroom
@sameroom-bot
[IFTTT]
New release from git tag: 0.5.1: Merge pull request #361 from alexbrillant/hotfix-warnings-warn-exception
New release from git tag: 0.5.1: Merge pull request #361 from alexbrillant/hotfix-warnings-warn-exception
Sameroom
@sameroom-bot
[IFTTT]
New release from git tag: 0.5.2
New release from git tag: 0.5.2
Sameroom
@sameroom-bot

[Guillaume Chevalier, Neuraxio] Hello @Everyone ! I'm presenting on Machine Learning's Design Patterns and Clean Code at the CAMDEA Digital Forum where I'll show some neat tricks.

You can register to the webinar here:
https://www.meetup.com/CAMDEA-Digital-Forum/events/272039548/
https://www.crowdcast.io/e/canada-aiml-data-science-16
Best,
Guillaume Chevalier

Sameroom
@sameroom-bot
[Adam Delgado, Neuraxio] Looking forward to your presentation. Thanks for the invitation to the slack channel :)
Sameroom
@sameroom-bot
[ajay vamsi, Neuraxio] hello everyone !
Sameroom
@sameroom-bot
[Guillaume Chevalier, Neuraxio] Reminder: I will present at the CAMDEA Digital Forum between approx 6:25pm- 6:45pm! Link in my previous comment above.
Sameroom
@sameroom-bot
[aihang, Neuraxio] :dancer: Just arrived!
Sameroom
@sameroom-bot
[Antoine Tremblay, Neuraxio] Hi, just presenting myself I'm Antoine Tremblay working at https://www.shutterstock.com I'm a principal engineer there working in the AI Eng team... basically my role is to assist the Data Science team by developing data , training and inference pipelines mainly related to Computer Vision scenarios.. some challenges:
• Scale: We don't use local disks except for non-mandatory caching
• Making sense of all the possible tools and ways to go about anything (The most difficult part imo)
• Performance and cost efficiency (on Cloud vs on Prem, CPU vs GPU intensive tasks)
We currently have a small pipeline framework I made but are looking at using neuralexe... main reasons are:
• Having a framework agnostic code
• Something we are not the only ones to dev, and can collaborate on
• You guys understand how to make good code which can by rare in the python / DS world unfortunately
[Antoine Tremblay, Neuraxio] These days I'm toying with the idea of having pipelines that could be powered by a suite of services potentially using https://developer.nvidia.com/nvidia-triton-inference-server since in a CV pipeline many steps are CPU intense while others are GPU and that having proper machines on AWS that can do both is not really efficient, I'm wondering if it would be a good arch to have transformers be stand alone in a grpc service for example and build service driven pipelines like:
Get Images -> Resize Images -> InceptionV2 -> Train Some model
[Antoine Tremblay, Neuraxio] Any ideas / references / comments welcome :)
Sameroom
@sameroom-bot
[Antoine Tremblay, Neuraxio] Oh and while I'm at it if you feel like this is something you'd like do professionally we're recurting
Sameroom
@sameroom-bot

[Guillaume Chevalier, Neuraxio] Hi @Antoine Tremblay ! I'm glad that you enjoyed our Clean Machine Learning training back in March, and that it is solving practical business problems for you and your team. Note that at Neuraxio, we can offer some coaching on the side of our other activities if you'd like us to prepare you a plan : Computer Vision is not our main focus, although I've personnaly worked on 5+ (if not 10) Computer Vision projects and that we have the capacity to offer you a detailed plan/design for your pipeline, that your internal ressources & employees could implement. You can write to me in private to request a few coaching hours as such.

In case you missed my last conference in mid-September, here is the recording link. Note that it is completely different than the Clean Machine Learning Training you attended in March:
https://www.youtube.com/watch?v=OFe223rUBY8

Sameroom
@sameroom-bot
[Antoine Tremblay, Neuraxio] Was there ever thought of having Neuralexe pipelines running as metaflow steps, or as kubeflow pipelines ?
Sameroom
@sameroom-bot
[Adam Delgado, Neuraxio] Hey Gang!
CAMDEA Digital Forum will have a booth at the Hackernest Tech Social event on Saturday 14th November from 2pm-4pm.
You can check out their website here: https://hackernest.com/
Hackernest is a not-for-profit technical networking and social event and has 16k+ members on their meetup group: https://www.meetup.com/hackernest/
Both Hackernest & CAMDEA Digital Forum are focused on adding value to the Canadian tech community in a not-for-profit model so it's a great partnership.
Perhaps you'd like to join us on Saturday?
Stay Healthy & Happy!
Sameroom
@sameroom-bot
[Antoine Tremblay, Neuraxio] Question for the Nauraxle lib... say I wanted a Step to not pass data to the next step and just end processing of that batch and pass to the next one ... what would be the proper way ?
Context is that I'm doing some feature extraction and want to skip processing if I find the feature already stored...