coffee_am

6 ms·

1.
by coffee_am 8mo ago
On the other side of the equation I've been spending much more time on code-review on an open source project I maintain, because developers are much more productive and I still code-review at the same speed. The real issue is that I ca
2.
by coffee_am 1y ago
Is there a public curated list of "good ips" to whitelist ?
3.
by coffee_am 1y ago
imho that is just silly ... I can see various ways censorship and freedom and common good at the same time. Actually, I can imagine different set ups where this could work... But then, you have to define these things. E.g.: freedom of perso
4.
by coffee_am 1y ago
Of course one can generalize using the colloquial "Here in Europe". And generalization is useful -- one cannot go into all the complexity and details all the time, at some point one has to summarize/generalize an argument. Ye
5.
by coffee_am 1y ago
Just for another data point, I took me 4 days to cook up a WASM front-end using Go for my otherwise command-line only Hive game: https://janpfeifer.github.io/hiveGo/www/hive/ Probably everything JS and DOM is
6.
by coffee_am 2y ago
Nice, but why create yet another language, that doesn't introduce no real gains over existing ones ? It's not even a research effort ... All modern languages will compile to WASM as well, that is not a new feature. I'm failin
7.
by coffee_am 2y ago
I think there is a misconception there. Go is as good a language as any for an ML framework. Better if you buy into it being better for its simplicity. It's clearly a worse if you factor in the ecosystem that already exists in Python,
8.
by coffee_am 3y ago
A few considerations come to mind: 1. The O(N^2*d) computation cost of the attention layers. For large graphs (millions of nodes) it's quickly too costly. And in some of the social network problems, the more data you feed the better is
9.
by coffee_am 3y ago
Thanks for the reply, but when you mean "fine-tuning" for the prompt, do you mean fine-tuning of the LLM encoder of the prompt right ? (The thing that transforms the prompt into a sequence of embeddings?) But that is not cheap&#x
10.
by coffee_am 3y ago
Noob question: when folks talk about fine-tuning LLM, do they usually fine-tune the encoder (of the prompt), the decoder (that generates the text) or both ?
11.
by coffee_am 4y ago
Sadly (but correctly) nothing is public, no one ever sees any data, it's a service. Pure FNN (feedforward NN) models, if I recall correctly is also ~30 to 40%. Since the server doesn't work for all types of data, and probably folk
12.
by coffee_am 4y ago
One way is to use categorical set splits [1] (proposed for categorical set inputs, but works for categorical features as well), used in TF-DF [1]. Greedy, and expensive to train (cheap inference though), but it gives great results. [1] htt
13.
by coffee_am 4y ago
I can't explain it, but I help maintain TensorFlow Decision Forests [1] and Yggdrasil Decision Forests [2], and in an AutoML system at work that trains models on lots of various users data, decision forest models gets selected as best
14.
by coffee_am 6y ago
DeepL translation "fork away" -> "Weggabelung" ... not much better :\