From the course: Large Language Models on AWS: Building and Deploying Open-Source LLMs
Unlock this course with a free trial
Join today to access over 24,600 courses taught by industry experts.
Python UV scripting
From the course: Large Language Models on AWS: Building and Deploying Open-Source LLMs
Python UV scripting
- [Instructor] One real problem with doing AI and ML is you have to deal with Python packaging. And for a long time it's been non-deterministic, slow, but, there is some help on the way. And this is a tool that could come in handy for a project you're working on for large language models, as you need to use Python somehow. First step, install UV. So we can see here, I do a curl to astral.sh here and boom, we get UV running. It's ridiculously fast. And one thing to point out, is if I go to which uv, ah, here we go. It's in my .local/bin. And if I do a du command on this, we say du - sh, here, home/Noah/.local/bin/uv, you can see it's tiny. So this is the advantage of RUST right? It's a RUST based tool that fixes the problem with Python packaging, which is a little bit ironic, is it's tiny, binary optimized, just incredibly fast. And if you want to run it, you can go right here. And you can see there's all kinds of cool options here. Now, the feature that I think is really interesting…
Contents
-
-
-
(Locked)
Implications of Amdahl’s law: A walkthrough4m 5s
-
(Locked)
Compiling llama.cpp demo4m 17s
-
(Locked)
GGUF file format3m 18s
-
(Locked)
Python UV scripting3m 55s
-
Python UV packaging overview1m 59s
-
(Locked)
Key concepts in llama.cpp walkthrough4m 37s
-
(Locked)
GGUF quantized llama.cpp end-to-end demo4m 3s
-
(Locked)
Llama.cpp on AWS G5 demo4m 20s
-
(Locked)
-