@OpenAI Gilfoyle: This thing is addressing problems that don’t exist. It’s solutionism at its worst. We are dumbing down machines that are inherently superior.
@thdxr The messenger mac app has a new update thats awful. To install the non-awful previous version I go to the messenger github repo and get the .rb file of the version I want. Paste it in ~/homebrew-local-casks/Casks, brew tap-new local/messenger and then install that 🤡
[1/*] If we take the chance that a tool (compiler, linker, batch, whatever) remains working for a particular codebase after one year as a given probability p, then the chance that build remains working after x years is p^xn, where n is the number of tools used in the build.
If you are just getting into ML, do yourself a favour and checkout @scikit_learn 's **FREE** MOOC (organised by @InriaLearnLab)
It covers:
- ML Concepts
- ML pipelines
- Selecting the best model
- Hyperparam tuning
- Linear models
- Decision trees
- Ensembles
- Evaluation
I was looking at the VarianceThreshold from the @scikit_learn docs, and it says it "Learn[s] empirical variances from X". I thought empirical variance means sample variance (i.e. we divide by n-1). But then I compared the variance with pandas' variance method and (1/x)