Author: sergii_kharlanov

  • STDs can be different

    std – article Innocent beginning¶ I was figuring out normalization and came across some code where the MNIST dataset was divided by 255 before normalization. I began to suspect, based on observations from this article: How ML models work with images? that this is incorrect. Not incorrect, but unnecessary, because normalization should “handle” this division…

  • Does the attempt we put on for this GMAT exam really helpful in later life?

    There was a period in my life when I quit my business and started my journey to an MBA.The main obstacle is usually the GMAT exam, which is really challenging.I like the saying that “GMAT is an Ironman for nerds”.One crucial part of my preparation was participating in the GMATClub site/forum, where people solve tasks,…

  • How ML models work with images?

    load_images_for_ML Intro¶ If you read chapter 4 from Practical Deep Learning for Coders and you have questions about all these numbers from images like what is $28*28$ or $784$ or why we divide these tensors with image data on 255 then this is a post for you Usual start of working with images¶ In [1]:…

  • RMSE / L2 normalization

    This is loss metric for machine learning models.Has two names: Computing RMSE To compute RMSE we need to Actual Predicted Error Powered error 10 20 -10 100 3 8 -5 25 6 1 5 25 Total powered error: 150Mean powered error: 50Root mean squared error: 7.07 Python code to compute RMSE – SciKit Python code…

  • MAE / L1 normalization

    This is loss metric for machine learning models.It has different names: Computing MAE To compute MAE we need to: Actual Predicted Error 10 20 -10 23 13 10 5 6 -1 Total absolute error: 21Mean absolute error: 7 Why absolute values? If we compute non absolute difference, then negative errors can cancel out positive errors:…