notpublic

10 ms·

31.
by notpublic 2y ago
Thanks for that link! There is another video posted 7 months ago from Seth where he interviews Ken Stanley, the creator of the NEAT algorithm. https://www.youtube.com/watch?v=5zg_5hg8Ydo
32.
by notpublic 2y ago
That demo is pretty impressive!
33.
by notpublic 2y ago
search for "Ultrasonic Mist Maker"...
34.
by notpublic 2y ago
Andrej Karpathy https://www.youtube.com/@AndrejKarpathy/videos Sasha Rush (an underrated creator who deserves more eyeballs) https://www.youtube.com/@srush_nlp/videos
35.
by notpublic 2y ago
> it's fun and simple Not sure if it was mine or my friends car, replacing one of the lights was neither fun or simple :). Some manufacturers really seem to prioritize style over repairability!
36.
by notpublic 2y ago
Llama-3.2-11B-Vision-Instruct does an excellent job extracting/answering questions from screenshots. It is even able to answer questions based on information buried inside a flowchart. How is this even possible??
37.
by notpublic 2y ago
Nice and Thanks! Was looking for something like this with minimal dependency. By the way, changing options updates browser history every time. You may want a debounce. Also, you could remove jQuery and make it even more leaner.
38.
by notpublic 2y ago
Presentation by the same author made 2 months back: https://www.youtube.com/watch?v=FYYZZVV5vlY
39.
by notpublic 2y ago
Have a used HP Prodesk 600 G3 with 16GB RAM running idle around ~12W. Bought it last year and its been running solid so far.
40.
by notpublic 2y ago
run it inside podman/docker
41.
by notpublic 2y ago
Having read that article, I have to say that is spot on!
42.
by notpublic 3y ago
Tim Dettmers QLoRA https://nips.cc/virtual/2023/83963 More here: https://nips.cc/virtual/2023/competition/66594
43.
by notpublic 3y ago
time-series data? Maybe include timescaledb
44.
by notpublic 3y ago
Elixir opentelemetry works quite well with Tempo. Tempo does the metric generation [1] and writes it to Prometheus. Tempo also does Service Graphs which works great with context propagation [2]. Btw, metric generation is not enabled in Temp
45.
by notpublic 3y ago
Video of the presentation: https://www.youtube.com/watch?v=QNwQ90hI874
46.
by notpublic 3y ago
https://github.com/github/copilot.vim https://github.com/huggingface/llm.nvim
47.
by notpublic 3y ago
In US as well..
48.
by notpublic 3y ago
source code: https://github.com/vllm-project/vllm
49.
by notpublic 3y ago
Indeed and thanks for link! It's been a while since I watched it.
50.
by notpublic 3y ago
Then you should watch the OG - 1979 Tinker Tailor Soldier Spy with Alec Guinness and Ian Richardson
51.
by notpublic 3y ago
I find it quite readable especially with nested logic. Here is an example: https://github.com/elixir-ecto/ecto/blob/master/lib/ecto/que... Edit: Just remembered another use case where function
52.
by notpublic 3y ago
If the model is stored in 32bits, it will be ~4x param size You can see the actual size from hugging face. For example, https://huggingface.co/WizardLM/WizardLM-7B-V1.0/tree/main Size with quantization, http
53.
by notpublic 3y ago
+1 Switched from sshfs to NFSv4+wireguard few years back. Works great!
54.
by notpublic 3y ago
As mentioned in the OP's blog post, checkout https://github.com/facebookresearch/llama-recipes.git . specifically files in ft_datasets directory. I am able to finetune meta-llama/Llama-2-13b-chat-hf on a 3090
55.
by notpublic 3y ago
Got it immediately after signing up through huggingface https://huggingface.co/meta-llama/Llama-2-7b-chat-hf
56.
by notpublic 3y ago
This video may be of interest to you: https://youtu.be/pQ0CvjAJXz4?t=1992 Bryan Hunter explains how they do it at HCA healthcare (not specifically LiveView process, but processes in general).
57.
by notpublic 3y ago
not an AI expert but from a talk I recently heard... if there is a mismatch in training data between the "teacher" LLM and "student" LLM, you risk teaching the student to hallucinate or to ignore information
58.
by notpublic 4y ago
"IMPORTANT: BY USING YOUR MAC PRODUCT, YOU ARE AGREEING TO BE BOUND BY THE TERMS OF THE APPLE 1 YEAR LIMITED WARRANTY (“WARRANTY”) AS SET OUT BELOW...." If the product is that reliable, why not offer a 3-5 year warranty?
59.
by notpublic 4y ago
Direct links to Pine64 products ... 128Mb Ox64 SBC - $8 USD https://pine64.com/product/128mb-ox64-sbc-available-on-decem... 16Mb Ox64 SBC - $6 USD https://pine64.com/product/16mb-ox64-sbc-available
60.
by notpublic 4y ago
Here is a good a talk from ElixirConf 22 https://www.youtube.com/watch?v=Y2Nr4dNu6hI "Chris Grainger (Amplified) delivers the Friday Keynote at ElixirConf US 2022. We switched our machine learning research, production

More ›