-
## Issue will be closed if:
> 1) You mention more than one algorithm. You can create a separate issue for each algorithm once the current one is completed.
> 2) You propose an algorithm that is …
-
Add support for Sliding Windows.
Sliding Windows are fixed-time windows determined by the record timestamps.
For example, calculating the average sum over the last 5 minutes requires sliding win…
-
-
### 🐛 Describe the bug
Hi AMD Team,
I am currently trying to improve Mistral performance on MI300X by attempting to use FlexAttention Sliding Attention but unfortunately it is running into numer…
-
Basic concept of Rate Shield is to implement different rate limiting strategies and integrate them with Redis. Than user can choose which strategy to apply and what rule to apply on particular API end…
-
Using ``.rolling().construct()`` will inject new dimensions into an array with ``sliding_window_view`` under the hood. NumPy is clever about this by arranging things to be a view of the underlying arr…
-
Based on what we've seen so far a strategy can be great for a day but once new data arrives the strategy may turn to crappy. 04-06-2024 strategy turned crappy 900 equity out of a 1000 for ARM but just…
-
### Is there an existing issue for this?
- [X] I have searched the existing issues
### Feature Description
Template for sliding window approach
1. Fixed Size Window
2. Variable Size Window
…
-
In Hugging Face "eager" Mistral implementation, a sliding window of size 2048 will mask 2049 tokens. This is also true for flash attention. In the current vLLM implementation a window of 2048 will mas…
caiom updated
3 weeks ago
-
### Initiative (Required)
Hacktoberfest 2024 🎃
### Is your proposal related to a problem? Please describe.
Finding the maximum product of the subarray that could be obtained from the input array wh…