-
See the final slides in https://cseweb.ucsd.edu/classes/sp17/cse168-a/CSE168_16_HighPerformance.pdf
![image](https://user-images.githubusercontent.com/26149045/209213539-7d88ba8d-955a-4c95-8c2f-9bf…
-
Currently our archive cache is very basic, there is no invalidation causing it to grow as the user installs packages.
We should look at a more sophisticated approach with size limits e.g an LRU ca…
-
**Describe the bug**
- Caching improves the speed and efficiency of github workflows - we currently cache quite aggressively.
- However, if there is for example a database migration in a PR, but t…
-
If I understand correctly, installing from git URL creates the following in the uv cache:
1. the git objects in `git-v0/db`
2. a git checkout in `git-v0/checkouts` (with hardlinks to the git objects…
-
Hi @frdel, thanks for your great work!
I found AutoMemory to be quite inefficient because it produces a summarization of the same memories at every step of the conversation.
### Suggestion
My…
-
**Why is it that when using a quantitative model for inference, the TTFT optimization is not obvious, but the overall inference efficiency is improved a lot? At the same time, the inference efficiency…
-
Currently both the [`backend`](https://github.com/grocy/grocy-docker/blob/22d6fa0c946aa5658d4305b3061919190557baeb/Containerfile-backend) and [`frontend`](https://github.com/grocy/grocy-docker/blob/22…
-
### 🚀 The feature, motivation and pitch
**Feature:** Add functionality to save and load models compiled with `torch.compile`.
**Motivation:** I'm working on a project with a large diffusion model …
-
We may need to support refreshing only part of the cache at a time, for efficiency, based on the outcome of #23. One approach here: limit the number of jobs done periodically, and sort jobs for the ne…
-
!pip install -U airllm
!pip install -U bitsandbytes
!pip install git+https://github.com/huggingface/transformers.git
!pip install git+https://github.com/huggingface/ac…