ImaCake

9 ms·

121.
by ImaCake 2y ago
> Unfortunately there's a lot of lag A great generalisation and understatement! Often looking like you are becoming more efficient is more important than actually being more efficient, e.g you need to impress investors. So you cut b
122.
by ImaCake 2y ago
This autocomplete seems about on par with github copilot. Do you also get options for prompting it on specific chunks of code and performing specific actions such as writing docs or editing existing code? All things that come standard with
123.
by ImaCake 2y ago
Not neural nets. CMIP models are largely dynamical models.
124.
by ImaCake 2y ago
But the system they have succeeded in optimises for people who can sell themselves well enough to get that funding. These people live and breathe selling themselves for funding. Every buzzword, sexy plot, and dynamic presentation has got th
125.
by ImaCake 2y ago
The corners of physics I have some contact with (climate/weather modelling and astrophysics) seem pretty dynamic to me. Each generation of CMIP models seems to be significantly better than the previous.
126.
by ImaCake 2y ago
Quality of Life Years, QALYs, is a commonly used term in epidemiology/biostatistics.
127.
by ImaCake 2y ago
I think Revelation Space does a great job creating a universe that squeezes out novel human cultures through the crushing vice of selection pressure. It’s every bit as daring as The Culture, just a different vibe!
128.
by ImaCake 2y ago
Make sure you enforce use of curly braces like all good languages
129.
by ImaCake 2y ago
A conference I helped organise had a lot of “AI” and “LLM” submissions. We had a blanket rule of ignoring all of them since they mostly sucked anyway. Meanwhile another conference I am going to has several “machine learning” talks which cou
130.
by ImaCake 2y ago
The article appears to be deliberately obfuscating the definition to keep the reader hooked for longer. The link to the article that describes a geofence is even more transparent about it.
131.
by ImaCake 2y ago
Smoking anything carries an increased risk of developing COPD. It’s not super important what plant it is.
132.
by ImaCake 2y ago
Agreed. Water has an LD50 and you can drown from drinking too much. But obviously we die without enough of it.
133.
by ImaCake 2y ago
No it is not. By not distinguishing consumption methods the authors likely gained a huge increase in sample size which helps them have more confidence in their statistics. There are trade offs here which are not obvious without understandin
134.
by ImaCake 2y ago
This is a terrible headline for an already over-editoralised paper. The entire study revolves around 2 people (twins) and thus no statistics can be performed. Statistics is careful to prune away hypothesises but I am happy to state that thi
135.
by ImaCake 2y ago
"Generative" is a great technical term for LLMs because it is what they do. But it a bad marketing term because it doesn't describe what they can do well. >object detection This comes from a different class of machine le
136.
by ImaCake 2y ago
Probably the kind of intuitive AI assistant you can find in a lot of science fiction works. A virtual assistant that can search the web for you, write/proofread an essay, analyse social situations, and enrich the right bits of your per
137.
by ImaCake 2y ago
Certainly no detectable trend in that data. But the accident frequency is so low that the random variation dominates and makes it impossible to distinguish any trend. What is demonstrably increasing is CAT, due to climate change. But consid
138.
by ImaCake 2y ago
Im sure that’s part of it. But you could say the same for using python or R over another language. Besides, someone who knows R well enough to write DplyR thought the situation was dire enough to write it. And there’s also data.table but th
139.
by ImaCake 2y ago
Maybe it is for you, but the success of Dplyr and ggplot suggests a lot of others disagree.
140.
by ImaCake 2y ago
You can use the bing copilot if you want your LLM to use references.
141.
by ImaCake 2y ago
I use LLMs for either code examples or explanations (which I can then verify) or for details on a topic I am not familiar with. Often much quicker than wading through the SEO chaff for the search equivalent.
142.
by ImaCake 2y ago
>ported to python are just objectively worse in that language This is absolutely the case. Dplyr syntax is much more intuitive for many use cases than Pandas or Polars equivalents. One thing I miss from RStudio is the Rmarkdown documents
143.
by ImaCake 2y ago
Except they kinda are? LLMs are just word models built from a corpus of the internet. There are examples of GPT3 regurgitating reddit comments in full given the right prompt. Certainly I find LLMs replace a lot of searches for me and googl
144.
by ImaCake 2y ago
Surely this has happened with this latest change? It got turned on for my work machine a few months ago and tried to upload some very large modelling files to the cloud. I have to keep that stuff in the root dir now, but most PC users don
145.
by ImaCake 2y ago
Yes, and that is for regulatory monitors. You might need one such monitor at a critical sensitive location and then get by with various e-sampler monitors elsewhere which kinda suck but are good enough to help build a picture of emissions.
146.
by ImaCake 2y ago
>per continuous monitoring for a broad range of chemicals would cost about $1 million per monitoring station This seems a bit high, closer to $100k installation cost for fairly basic real time regulatory monitors seems about right. CEMS
147.
by ImaCake 2y ago
I think these guidelines are also a political problem. They need to balance risk, communication, and what is feasible given the resources available and threat level. That’s not an easy thing to do. Still I am deeply skeptical that the peopl
148.
by ImaCake 2y ago
I think this says more about the guidance level than the soil quality. In fact I think it makes things worse rather than better. Continually ratcheting up the safety level to spur continuous improvement is great in principle. But what I thi
149.
by ImaCake 2y ago
To those who can’t see why you need to distinguish between crawlers and user agents the reason is accessibility . Some people are blind, others have physical disabilities, some of us have astigmatisms or ADHD and can’t use badly designed a
150.
by ImaCake 2y ago
Him making this joke is also a product of his expertise. Like, it takes at least some experience to have the insight and then be able to communicate it in a pithy way.

More ›