Where communities thrive

  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
Repo info
    Sandeep Aswathnarayana
    @ssgoob I sent an email to Team@fullstackdeeplearning.com with a few queries on July 10. I understand how busy you probably are. Would you mind answering them at your convenience?
    Does anyone happen to know if the recent ICML presentation videos are open-sourced? - I am interested in Snorkel AI's.
    Mauricio Oliveto
    Is it off-topic to ask if anyone knows about a computer vision dataset for detecting COVID-19 related human poses?
    1 reply
    Is anyone using the Thinc library? I was thinking about giving it a go on my next project -- love the idea of stronger type checking. https://thinc.ai/
    2 replies
    Servando Torres
    I'm wondering if Coursera and Udacity course still hot within your industry/country. With the massive democratization of information, perhaps there are different trending tools in certain areas
    Josh Tobin
    Interesting article: https://blog.kubeflow.org/mlops/
    Vinicius Ribeiro Machado da Silva
    Hey @sergeyk , Vinicius from last FSDL 2019. Is there any possibility to have access to the videos and notebook from this class here? https://docs.google.com/document/d/e/2PACX-1vRVecewhTEhIVcXXOf0iD0Okkm-DvgcBj7nHQFPgN9v1kbQ7gCkCn-VUE4Lz-p36pml0wIIQmI3begD/pub
    Danielh Francisco Mendez Carranza
    Hi all, I built a codebase & GitHub template based on the FSDL labs. Use it as a starting point to get your Machine Learning project into production. I hope is useful for all! https://github.com/DanielhCarranza/ml-production-template
    2 replies
    Hi everyone
    Andrea Panizza
    @josh_tobin__twitter 不不不不不
    Carlos Leyson
    Hi guys, I am going through the bootcamp and I am currently in lab1 but I am having issued when trying to train the model.
    6 replies
    James Le

    Great post from a16z on long-tailed AI problems: https://a16z.com/2020/08/12/taming-the-tail-adventures-in-improving-ai-economics/

    My favorite quotes:
    1 - Long-tailed distributions are also extremely common in machine learning, reflecting the state of the real world and typical data collection practices.

    2 - The long tail can contribute to high variable costs beyond infrastructure. Even worse, AI businesses working on long-tailed problems can actually show diseconomies of scale meaning the economics get worse over time relative to competitors.

    3 - ML is not a religion, but science, engineering, and a little art.

    4 - If you are working on a long-tail problem, its critical to determine the degree of consistency across customers, regions, segments, and other user cohorts. If the overlap is high, its likely you can serve most of your users with a global model (or ensemble).

    5 - In "componentizing", each model addresses a global slice of data rather than a particular customer, for instance and that the sub-problems are relatively bounded and easy to reason about.

    6 - Over time, models that share similar functionality can be joined together with a common trunk to reduce complexity. The goal is to make the trunk models as thick as possible (i.e. doing most of the work) while making the task-specific branch models as thin as possible without sacrificing accuracy

    James Le
    Such a comprehensive overview of the state of machine learning model servers in production, as well as the core components that constitute full-featured systems: https://anyscale.com/blog/heres-what-you-need-to-look-for-in-a-model-server-to-build-ml-powered-services/
    Danielh Francisco Mendez Carranza
    Such a great talk about building safe, reliable, and robust models to dataset shifts. They use a proactive approach (models that are stable to any anticipated shifts) with DAGs were they combined prior knowledge with structure learning algorithms that partially learn the graph from data. https://www.youtube.com/watch?v=0EP_wR0YT4A&t=2140s
    James Le

    "An engineering team that built their own ML Platform from the ground up, flawed as it may be, will attribute more value to it than if they just bought something out-of-the-box from a vendor. They give it a fancy name, write blog posts about it, and everyone gets promoted."


    Chester Chen
    which one is our main chat app ? Gitter, Discord now ?
    Servando Torres
    Cites the FSDL bootcamp!
    1 reply
    Andrea Panizza
    Oh, hey, nice piece!
    Danielh Francisco Mendez Carranza
    A well explained and informative paper about the key areas of success and challenges of Monitoring and explainability of ML models in production (Monitoring model performance, Monitoring metrics related to incoming data, outlier and shift detection, explainability of models, etc.) https://arxiv.org/pdf/2007.06299.pdf
    1 reply
    Chester Chen
    quick question for the group: Can someone provide real-world use cases that real-time ML model training ?
    1 reply
    Guillaume "Vermeille" Sanchez
    Hi guys! I'm following the course and am a ML eng for a small company
    I was wondering if you guys had any pointers to the toolchain / stack that would make deploying monitoring and dashboards easy for a model served with flask?
    James Le
    DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective: https://github.com/microsoft/DeepSpeed
    hi Guys .i am new here, how could i enroll in this course? please help
    James Le
    Shaul Solomon
    Can anyone inform me the order of Watching the Videos & Doing the Labs?
    Serg Masis
    Hello all! My company (Syngenta) is looking for a Deep Learning Scientist embedded in the Seeds R&D department working on next generation of seeds development optimized for climate and pest resilience. Theres genetics, computer vision and even NLP involved. If this interests you and you can work in the US, I know the hiring manager and can put in a good word. https://syngenta.taleo.net/careersection/unitedstates/jobdetail.ftl?job=18013166
    I got this error when running pip-sync requirements.txt requirements-dev.txt
    can someone please me help fix this?
    Max 云
    Hey yall! My company, KUNGFU.AI, an AI consultancy, is hiring engineers and an intern - check us out. Great people at the company and I'm an FSDL alum, so lmk if u apply! https://www.kungfu.ai/about/careers/
    Servando Torres
    Hello! Does anyone know good resources on Ml / A.I tool monetization?
    Katimus Prime
    Shuang Song
    Glad to join this course
    Rancy Chepchirchir
    Hello.. happy to be here
    Sandeep Aswathnarayana
    Hello all! I understand there is a fee attached for normal registrations for the FSDL course. I received an email of acceptance for the Spring 2021 course today. Do I still need to pay for the course and certification to enroll?
    Jean-Marie Prigent
    Hi, I would like to know if there is any Security content aspect in FSDL Spring 2021 new course. I work in Banking ecosystems.
    2 replies
    Hi, everyone, great to join the community and I registered for the Spring 2021 course. How much time is expected per week to complete the course?
    Rahul Thakkar


    I have registered and paid for the upcoming spring-21 Berkeley offering of the course but I havent received any communication other than a Stripe receipt. And a few week later I got another email asking again if I want to register for the course. Did anyone get any communication after the payment from the FSDL team? I want to make sure that I am registered.

    I also see discrepancy in the schedule mentioned on the course page (supposed to start on week of Jan 19) and in the email (supposed to start on week of March 1). I had reached out to the FSDL team around a week back on a support email mentioned in the earlier emails but havent got any reply so any pointers would be appreciated. Thank you!

    Hi, I'm an undergrad student at Berkeley and I've filled out the application form for the spring-21 offering of FSDL. In its previous offerings, it seems that the course contents can be fit in a three-day bootcamp. Since now it is offered as a full-semester 3-day course, I wonder how much new/updated content are added into it? Does this have a big impact on the syllabus or course goals?
    • 3-unit instead of 3-day in the second line :)
    Ashik Shaffi

    Hey guys, I have little experience working in the area of deep learning before getting fully into it. Well, I love the idea of Data Science to be end - to - end who can actually deliver values too. Will it be fine taking this course at first and getting to know about the complete workflow of Deep learning end - to -end (or) work on deep learning for some time and jump back to this course?

    I am a little confused about how should I take this course, any suggestions? Thank ya!

    Aman Kumar Nayak
    Hey folks, is there any open slack or gitter chanel for March 2021 course for non paying learners ?
    as officially there is not
    Matthew Chung
    Hi, quick question. Is the content here https://course.fullstackdeeplearning.com/ the same as what will be in the March 2021 course?
    2 replies