-
### What is your issue?
A common use case for us working with xarray + zarr for ML is sampling chunks of data from multiple zarr files. I have noticed that the time taken to load a sample differs sig…
dfulu updated
1 month ago
-
At the moment we are doing experiments on mock datasets of 200 supernovae. What would it take to repeat these on 100,000 supernovae? Some benchmarking tests could be interesting - and let us know if w…
-
This issues describes the high level directions that "create LLM Engine V1". We want the design to be as transparent as possible and created this issue to track progress and solicit feedback.
Goal…
-
See PR #32 (thanks @ahmedtadde for the suggestion!)
Currently, we apply parsing tools each time we load or sample data. We could cache the output of `self._apply_parsing_tools()` for the entire da…
-
https://github.com/ggerganov/llama.cpp/pull/2926
https://github.com/ggerganov/llama.cpp/pull/3624
https://github.com/ggerganov/llama.cpp/pull/5625
**Feature request**
1. Implement speculative …
-
DPM adaptive:
` File "/content/gdrive/MyDrive/sd/stablediffusion/src/k-diffusion/k_diffusion/sampling.py", line 496, in sample_dpm_adaptive`
` if sigma_min
-
Hello!
A few quick questions. All of them are brainfood about interface improvements.
- how to properly perform fast seeking? sidplayfp::play with null buffer just renders single sample. Rendering…
-
Hey!
This is not an issue but rather a suggestion :).
From what I understand, you walk on a graph by randomly hopping from one node to another. You do this via a roulette wheel method that runs …
-
The LTC2325-16 is Quad, 16-Bit, 5Msps/Ch Simultaneous Sampling ADC.
The LTC2325-16 has a high speed SPI-compatible serial interface that supports CMOS or LVDS. The fast 5Msps per channel throughput.
-
### Anything you want to discuss about vllm.
Following is a little piece of code to extract embeddings from a certain layer of LLM:
```python
def process_row(prompt: str, model, tokenizer, layers_t…