Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Machine Learning and Ketosis

github.com

Post reply on HN
Print view View original

241–250 of 309 posts

Re: Machine Learning and Ketosis

#241

Post by philip1209 »

MicroBerto wrote: post #106
philip1209 wrote: post #5

Earlier quoted context omitted.

I'm in ketosis. I can tell when the glycogen in my liver depletes. I hit a wall and can't keep going. I find that it has to do with intensity - for low intensity exercise, I can go forever. When I have to exert more (like running up an SF hill), it depletes me more. To fix this, you more or less have to keep intensity lower. I think that's why it works well for bicyclists who can change gears. Good news is that, afte…

Look into BHB supplements. We're doing a lot of research on them and they may be just what you're looking for.

Yes. I have MCT oil every day, but haven't experimented with exogenous ketones.

Re: Machine Learning and Ketosis

#242

Post by ariel-faigon »

aab0 wrote: post #231
ariel-faigon wrote: post #76

Earlier quoted context omitted.

Thanks so much for all the excellent comments. There was definitely an over-fit with 4-passes. No more. I've updated the Makefile to run only one pass, changed the options so it runs with older-version vw, Fixed misspellings of 'gioza', removed 'mayo' which found itself on the wrong side because it appeared only twice and always alongside the bun and regenerated the chart. All the main conclusions remain intact. In t…

I don't think it matters whether you run 4 passes or 1 pass, it's still going to overfit. You can run an online linear regression in a single pass too, but that doesn't magick away the uncertainties. The results are still going to be garbage, and any effects you get are due to your health-consciousness and not any specific dietary choices you make (how could it be, when the data is so weak and noisy that each item ca…

Thanks so much. Your comments are really helpful.

I realized early on that the data is hopelessly noisy, due to the small daily changes and the scales resolution so rather than trying to build a perfect model to gauge the variable importance of each and every kind of food, I focused on the few days when weight change was more significant hoping I could detect some signal in those, and extrapolate and further explore from that. That's why I sorted the data-set by abs(delta) and that's what consistently pointed me towards sleep/fasting as the #1 factor. I do agree that the full list/model is garbage in the sense that probably 80% or so of it is woefully inaccurate/flipped, noisy, overfitted etc. The main point was to lead me in the right direction by looking at the big picture and what stood out.

And what stood out were 2 things 1) sleep (fasting duration), and 2) fat vs carbs. I think everything else should be ignored. I think we're in total agreement on this point.

Does this sound more sensible to you?

Re: Machine Learning and Ketosis

#243

Post by notyourloops »

lucidguppy wrote: post #193

Carbohydrates do not make you fat. I am eating a high carb vegan diet and have lost 63 pounds since last October. My weight loss: http://imgur.com/a/cGb4X I do not restrict the amount of food I eat. I snack and have big meals. Carbs do not make you fat. Eating high caloric density foods makes you fat. More resources: http://nutritionfacts.org/ https://www.youtube.com/watch?v=d5wfMNNr3ak http://www.forksoverknives.com…

Carbohydrates make me fat. Maybe not you, but definitely me. There is no way around that and I've made peace with it.

Re: Machine Learning and Ketosis

#244

Post by notyourloops »

WillPostForFood wrote: post #143
ryeguy wrote: post #138

This is delightfully nerdy and overly analytical, but I can't believe someone would go to the extent of creating this yet never think to look for counterarguments against their dieting regimen. Doing this would have given him the answer much sooner. Trying to figure out how the consumption of certain foods correlates with weight gain/loss is a waste of time because it's simply the caloric content, not the macronutrit…

Food does have properties (not arbitrary) that make them good or bad for weight loss. I agree, calories count. But if you eat food that makes you hungry, it is harder to cut calories. The reason low-carb high fat diets work is that you feel full on fewer calories.

All calories are not equal. Ask a diabetic to chow down on carbs. The body clearly doesn't treat all the macronutrients as interchangeable units of energy. To say, "a calorie is a calorie" helps none and allows people to behave as thought-terminating parrots.

Re: Machine Learning and Ketosis

#245

Post by notyourloops »

When I ditch the carbs and eliminate grain from my diet, there's no denying that I look better and feel better. I've cycled on and off several times now and there's no way to get around it: The carbs and grain have to go and they have to go forever.

If you can eat them, no problem. However, there's many of us that can't and we're going to talk about how to operate with that as baseline.

Re: Machine Learning and Ketosis

#246

Post by kmm »

benkuhn wrote: post #15

It's astonishing (and really awesome) that they were able to extract such a strong signal from daily weight swings! This makes me a lot more optimistic about the possibilities of quantified-self-type stuff. Three things I'm really curious about: - How significant are the estimates of lifestyle factors? Do you have p-values? If you bootstrap resample, how much do the rankings change at the extremes? - How much cogniti…

> - It looks like the predicted delta from going from a "nosleep" day to a "sleep" days is about 1.4 pounds (sleep coef minus nosleep coef). That seems fishy, or at least like it will stop working fairly soon because you can't actually lose 1.4 pounds/day sustainably. Is it possible there's something weird going on with the data or those variables don't have the obvious meanings?

That weight loss doesn't add up in the long term. You lose a lot of weight during your sleep both due to evaporation of water and by burning glucose/fat which you exhale. Obviously longer sleep translates into more water lost and more glucose/fat burned, but it will quickly be compensated by the body taking that water back up again and a larger appetite. You can't consistently lose water weight.

After a night of drinking, I can lose up to 3 kilos due to dehydration. But of course, I've probably gained weight due to the calories in the booze.

Re: Machine Learning and Ketosis

#247

Post by lucb1e »

onetwotree wrote: post #92

I'd really like to see this kind of analysis applied to a larger sample. If someone were to put together a little app to make this sort of data collection trivial and upload it to a central (preferably cheap) repository, would people be interested in contributing their data in exchange for analysis? Also, since this is HN, is anyone interested in building something like this? I've got an hour or two a week of spare t…

> would people be interested in contributing their data in exchange for analysis?

I am not overweight but I would certainly not mind doing this if it helps others.

Re: Machine Learning and Ketosis

#248

Post by ariel-faigon »

elo_ wrote: post #203

I collect some of my own data in 1/0 form in regard to whether I was productive the day before as well as a few other things (I have a google form for myself that I fill out each morning). I only have a month of data but here is what it spat out: FeatureName HashVal MinVal MaxVal Weight RelScore ^dreams 24546 0.00 1.00 +0.0705 47.11% ^shower 215555 0.00 1.00 -0.0239 -15.96% ^exercise 190069 0.00 1.00 -0.0350 -23.41%…

This is pretty hilarious. Thanks for sharing. It may be showing how much noise there is in a random data-set with too few data-points to draw statistically significant conclusions. It might also mean that you're on the right track and now that you have the tool, you can keep going until you find real enlightenment.

I really like your comment because you're one of the very few that actually tried the idea. Thats the #1 reason I put this on github, to help others conduct their own experiments on whatever they care about. If you ignore all my weight-loss journey story and data and just use some of these ideas to improve your own life, it was all worth it for me.

Re: Machine Learning and Ketosis

#249

Post by msoto »

gtrubetskoy wrote: post #70
sliverstorm wrote: post #64

Earlier quoted context omitted.

glucose is the only fuel Can you expand? Like, you thought that cells only metabolize glucose, and dietary fats & proteins are converted to glucose? Or...?

> you thought that cells only metabolize glucose, and dietary fats & proteins are converted to glucose Pretty much. I don't know where that may have come from, but I suspect I'm not the only one who naively thinks that this is the case. Somehow "ketone bodies" aren't mentioned in dietary conversations nearly as frequently as glucose.

Here is a detailed explanation: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2129159/

From the abstract:

During very low carbohydrate intake, the regulated and controlled production of ketone bodies causes a harmless physiological state known as dietary ketosis. Ketone bodies flow from the liver to extra-hepatic tissues (e.g., brain) for use as a fuel; this spares glucose metabolism via a mechanism similar to the sparing of glucose by oxidation of fatty acids as an alternative fuel. In comparison with glucose, the ketone bodies are actually a very good respiratory fuel. Indeed, there is no clear requirement for dietary carbohydrates for human adults. [...] Contrary to popular belief, insulin is not needed for glucose uptake and utilization in man.

Re: Machine Learning and Ketosis

#250

Post by splintercell »

Borkdude wrote: post #215

Warning: unpopular opinion here. Long term low carb dieting and ketosis is not without risk. Yes, you will see weight loss, but what about other health markers? Check out some of the information here: http://nutritionfacts.org/topics/low-carb-diets/ What has worked for me was a plant based diet without added oil and processed foods: https://medium.com/@borkdude/tl-dr-of-my-long-term-weight-lo... Been doing that for f…

For once I want someone who isn't morally opposed to eating animals tell me that low carb diets are bad.

Vegans and animal lovers don't like low-carb-high-fat diets because it requires eating a lot of animals, so I find it hard to trust them.

Post reply on HN