Oh its an ancient skit.

Rocks missing.

Better get in the hurricane bunker.

The weather rock.

Without a political path forward we’re not given another choice.

God I would be so happy if this piece of shit lost her primary.

I’m not surprised by #1 or #8, but I’m pleasantly surprised by the fact that Leo Breimans paper on Random Forest is still so high up there.

The algorithm is great in its simplicity and it’s mind if mind-blowing how a collective of mostly poor predictors can be used to create an ensemble that is generally predictive. It’s also so much less data intensive and energy intensive to train than a deep learning algorithm.

You only look once.

I spent a summer canning tomatoes to realize I don’t really use that many cans of tomatoes. I’ll use fresh tomatoes, but we literally had to change our diet because we had canned enough tomatoes that we had to re-arrange the kitchen. We had tomatoes in cans for literally years after that summer.

#1 tip for starting gardeners: take a week or two and actually write down what the hell you actually eat that is a vegetable, and grow that. I’m not saying don’t branch out and try new things, but focus on serving yourself and growing things you actually eat. If you don’t eat like… 20+ tomatoes a week, you probably don’t need more than 1 or 2 tomato plants, if even that.

You’ll be way more successful/ happy/ satisfied/ likely to continue or advance as a gardener if it doesn’t feel like a chore and its serving you.

Primary every last one.

Cross validation is a way of calculating the likely uncertainty of any model (it doesn’t have to be a machine learning model).

A common cross validation approach is LOOCV (leave one out cross validation), for small datasets. Another is K-folds cross validation. In any case, the basics is to leave out “some amount” of your training data, totally removed from the training process, then you train your model, then you validate it on the trained model. You then repeat this process over the k-folds or each unit of your training data to create a valid uncertainty.

So a few things. First, this a standard approach in machine learning, because once you get stop making the assumptions of frequentism (and you probably should), you no longer get things like uncertainty for free, because the assumptions aren’t met.

In some approaches in machine learning, this is necessary because there really isn’t a tractable way to get uncertainty from the model (although in others, like random forest, you get cross validation for free).

Cross validation is great because you really don’t need to understand anything about the model itself; you just implement the validation strategy and you get a valid answer for the model uncertainty.

Good stories, however they are told, have a beginning, a middle, and most importantly, an end.

Signal is great for this. Ive written whole chapters this way.

Huh. 90% of the time I’m like “this is a bad movie”

Its the lead folks

Good luck surviving on tomatoes.

Any blue will do amirite?

Supreme Court just kicked this one back to the people.