13 Matching Annotations
  1. Nov 2021
    1. It is important to scale features before training a neural network. Normalization is a common way of doing this scaling: subtract the mean and divide by the standard deviation of each feature. The mean and standard deviation should only be computed using the training data so that the models have no access to the values in the validation and test sets. It's also arguable that the model shouldn't have access to future values in the training set when training, and that this normalization should be done using moving averages.

      moving average to avoid data leak

    2. You'll use a (70%, 20%, 10%) split for the training, validation, and test sets. Note the data is not being randomly shuffled before splitting. This is for two reasons: It ensures that chopping the data into windows of consecutive samples is still possible. It ensures that the validation/test results are more realistic, being evaluated on the data collected after the model was trained.

      Train, Validation, Test: 0.7, 0.2, 0.1

  2. Aug 2021
    1. It’s possible to divide data analysis into two camps: hypothesis generation and hypothesis confirmation (sometimes called confirmatory analysis). The focus of this book is unabashedly on hypothesis generation, or data exploration. Here you’ll look deeply at the data and, in combination with your subject knowledge, generate many interesting hypotheses to help explain why the data behaves the way it does. You evaluate the hypotheses informally, using your scepticism to challenge the data in multiple ways.
  3. Jul 2021
  4. Sep 2019
  5. Oct 2016
  6. Dec 2015
    1. verify that new software can be legally loaded into a device to meet these requirements

      And this is the required means, that the router vendors prevent loading of software that does not meet the desired ends. Previous documents instead specified that DD-WRT not be loaded.

      The FCC document is no longer available: please see http://web.archive.org/web/20150803065407/https://apps.fcc.gov/kdb/GetAttachment.html?id=1UiSJRK869RsyQddPi5hpw%3D%3D&desc=594280%20D02%20U-NII%20Device%20Security%20v01r02&tracking_number=39498

      It is cited in https://via.hypothes.is/http://www.wired.com/2015/09/hey-fcc-dont-lock-wi-fi-routers/

    2. The software must prevent the user

      This, however, changes the discussion to make the user or operator, rather than a third party, unable to operate the router outside of the legal limits.

    3. Device Security

      This is the beginning of the discussion of interest in the FCC's proposal which would ban open source operating systems.

      It should be noted that this is not proposed by the FCC as a ban as such, but rather an effort to keep the devices in question operating within the law and regulations.

      Our concern is it changes the regulations to make it difficult or impossible to use open source or free software operating systems, and that initially it named DD-WRT as an operating system that vendors were to required to prevent from use.

      The document mentioning DD-WRT has subsequently been amended, For the relevant copy, see the Internet Archive at http://web.archive.org/web/20150803065407/https://apps.fcc.gov/kdb/GetAttachment.html?id=1UiSJRK869RsyQddPi5hpw%3D%3D&desc=594280%20D02%20U-NII%20Device%20Security%20v01r02&tracking_number=39498 at the bottom of page 2

    4. third parties

      The common sense of "third party" is a person other than the vendor and purchaser of some thing.

    1. DD-WRT

      The citation of DD-WRT The linked document is no longer available at the FCC link address.