Close Menu
    Trending
    • What health care providers actually want from AI
    • Alibaba har lanserat Qwen-Image-Edit en AI-bildbehandlingsverktyg som öppenkällkod
    • Can an AI doppelgänger help me do my job?
    • Therapists are secretly using ChatGPT during sessions. Clients are triggered.
    • Anthropic testar ett AI-webbläsartillägg för Chrome
    • A Practical Blueprint for AI Document Classification
    • Top Priorities for Shared Services and GBS Leaders for 2026
    • The Generalist: The New All-Around Type of Data Professional?
    ProfitlyAI
    • Home
    • Latest News
    • AI Technology
    • Latest AI Innovations
    • AI Tools & Technologies
    • Artificial Intelligence
    ProfitlyAI
    Home » The Machine Learning Lessons I’ve Learned This Month
    Artificial Intelligence

    The Machine Learning Lessons I’ve Learned This Month

    ProfitlyAIBy ProfitlyAIAugust 31, 2025No Comments6 Mins Read
    Share Facebook Twitter Pinterest LinkedIn Tumblr Reddit Telegram Email
    Share
    Facebook Twitter LinkedIn Pinterest Email


    in machine studying are the identical.

    Coding, ready for outcomes, decoding them, returning again to coding. Plus, some intermediate displays of 1’s progress. However, issues principally being the identical doesn’t imply that there’s nothing to be taught. Fairly quite the opposite! Two to 3 years in the past, I began a every day behavior of writing down classes that I realized from my ML work. In wanting again via a few of the classes from this month, I discovered three sensible classes that stand out:

    1. Maintain logging easy
    2. Use an experimental pocket book
    3. Maintain in a single day runs in thoughts

    Maintain logging easy

    For years, I used Weights & Biases (W&B)* as my go-to experiment logger. Actually, I’ve as soon as been within the prime 5% of all energetic customers. The stats in under determine inform me that, at the moment, I’ve skilled near 25000 fashions, used a cumulative 5000 hours of compute, and did greater than 500 hyperparameter searches. I used it for papers, for giant tasks like climate prediction with massive datasets, and for monitoring numerous small-scale experiments.

    My as soon as upon a time stats of utilizing W&B for experiment logging. Picture by the writer.

    And W&B actually is a good instrument: if you’d like lovely dashboards and are collaborating** with a workforce, W&B shines. And, till not too long ago, whereas reconstructing information from skilled neural networks, I ran a number of hyperparameter sweeps and W&B’s visualization capabilities have been invaluable. I may instantly evaluate reconstructions throughout runs.

    However I noticed that for many of my analysis tasks, W&B was overkill. I not often revisited particular person runs, and as soon as a undertaking was performed, the logs simply sat there, and I did nothing with them ever after. Once I then refactored the talked about information reconstruction undertaking, I thus explicitly eliminated the W&B integration. Not as a result of something was incorrect with it, however as a result of it wasn’t vital.

    Now, my setup is far easier. I simply log chosen metrics to CSV and textual content information, writing on to disk. For hyperparameter searches, I depend on Optuna. Not even the distributed model with a central server — simply native Optuna, saving examine states to a pickle file. If one thing crashes, I reload and proceed. Pragmatic and enough (for my use circumstances).

    The important thing perception right here is that this: logging will not be the work. It’s a assist system. Spending 99% of your time deciding on what you need to log — gradients? weights? distributions? and at which frequency? — can simply distract you from the precise analysis. For me, easy, native logging covers all wants, with minimal setup effort.

    Preserve experimental lab notebooks

    In December 1939, William Shockley wrote down an thought into his lab pocket book: substitute vacuum tubes with semiconductors. Roughly 20 years later, Shockley and two colleagues at Bell Labs have been awarded Nobel Prizes for the invention of the fashionable transistor.

    Whereas most of us aren’t writing Nobel-worthy entries into our notebooks, we are able to nonetheless be taught from the precept. Granted, in machine studying, our laboraties don’t have chemical substances or take a look at tubes, as all of us envision once we take into consideration a laboratory. As a substitute, our labs usually are our computer systems; the identical system that I exploit to put in writing these traces has skilled numerous fashions through the years. And these labs are inherently portably, particularly once we are growing remotely on high-performance compute clusters. Even higher, because of highly-skilled administrative stuff, these clusters are working 24/7 — so there’s at all times time to run an experiment!

    However, the query is, which experiment? Right here, a former colleague launched me to the thought of mainting a lab pocket book, and recently I’ve returned to it within the easiest kind attainable. Earlier than beginning long-running experiments, I write down:

    what I’m testing, and why I’m testing it.

    Then, after I come again later — often the subsequent morning — I can instantly see which ends are prepared and what I had hoped to be taught. It’s easy, but it surely adjustments the workflow. As a substitute of simply “rerun till it really works,” these devoted experiments turn into a part of a documented suggestions loop. Failures are simpler to interpret. Successes are simpler to duplicate.

    Run experiments in a single day

    That’s a small, however painful classes that I (re-)realized this month.

    On a Friday night, I found a bug that may have an effect on my experiment outcomes. I patched it and reran the experiments to validate. By Saturday morning, the runs had completed — however after I inspected the outcomes, I noticed I had forgotten to incorporate a key ablation. Which meant … one other full day of ready.

    In ML, in a single day time is valuable. For us programmers, it’s relaxation. For our experiments, it’s work. If we don’t have an experiment working whereas we sleep, we’re successfully losing free compute cycles.

    That doesn’t imply it is best to run experiments only for the sake of it. However at any time when there’s a significant one to launch, beginning them within the night is the proper time. Clusters are sometimes under-utilized and assets are extra rapidly obtainable, and — most significantly — you’ll have outcomes to analyse the subsequent morning.

    A easy trick is to plan this intentionally. As Cal Newport mentions in his e-book “Deep Work”, good workdays begin the night time earlier than. If you already know tomorrow’s duties at present, you’ll be able to arrange the precise experiments in time.


    * That ain’t bashing W&B (it could have been the identical with, e.g., MLFlow), however fairly asking customers to guage what their undertaking objectives are, after which spend the vast majority of time on pursuing that objectives with utmost focus.

    ** Footnote: mere collaborating is in my eyes not sufficient to warrant utilizing such shared dashboards. You’ll want to achieve extra insights from such shared instruments than the time spent setting them up.



    Source link

    Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
    Previous ArticleUnderstanding Matrices | Part 4: Matrix Inverse
    Next Article How to Develop a Bilingual Voice Assistant
    ProfitlyAI
    • Website

    Related Posts

    Artificial Intelligence

    The Generalist: The New All-Around Type of Data Professional?

    September 1, 2025
    Artificial Intelligence

    How to Develop a Bilingual Voice Assistant

    August 31, 2025
    Artificial Intelligence

    Understanding Matrices | Part 4: Matrix Inverse

    August 30, 2025
    Add A Comment
    Leave A Reply Cancel Reply

    Top Posts

    Synthetic data in healthcare: Definition, Benefits, and Challenges

    April 9, 2025

    Exploring RAFT: The Future of AI with Retrieval-Augmented Fine-Tuning

    April 4, 2025

    DeepMinds 145-sidiga rapport om säkerhetsstrategi för AGI

    April 4, 2025

    In-House or Outsourced Data Annotation – Which Gives Better AI Results?

    April 3, 2025

    Exploring Prompt Learning: Using English Feedback to Optimize LLM Systems

    July 16, 2025
    Categories
    • AI Technology
    • AI Tools & Technologies
    • Artificial Intelligence
    • Latest AI Innovations
    • Latest News
    Most Popular

    Generative AI is reshaping South Korea’s webcomics industry

    April 22, 2025

    AI comes for the job market, security and prosperity: The Debrief

    August 27, 2025

    BBC Uses AI to Resurrect Agatha Christie as Your Personal Writing Coach

    May 1, 2025
    Our Picks

    What health care providers actually want from AI

    September 2, 2025

    Alibaba har lanserat Qwen-Image-Edit en AI-bildbehandlingsverktyg som öppenkällkod

    September 2, 2025

    Can an AI doppelgänger help me do my job?

    September 2, 2025
    Categories
    • AI Technology
    • AI Tools & Technologies
    • Artificial Intelligence
    • Latest AI Innovations
    • Latest News
    • Privacy Policy
    • Disclaimer
    • Terms and Conditions
    • About us
    • Contact us
    Copyright © 2025 ProfitlyAI All Rights Reserved.

    Type above and press Enter to search. Press Esc to cancel.