krzysztof/landing_page
1
0
Fork 0
mirror of https://github.com/qdrant/landing_page.git synced 2026-10-04 18:38:30 +02:00
Code Issues Packages Projects Releases Wiki Activity

added link to the release

Browse Source
This commit is contained in:
Евгения Суходольская
2025-05-13 18:13:37 +02:00
parent 9c2d9272ca
commit 47e8db525d
1 changed files with 7 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
qdrant-landing/content/articles/miniCOIL.md
+7 -2
View File
@@ -232,8 +232,13 @@ Here are the specific characteristics of the miniCOIL model we trained based on
| **Training Parameters** | **Epochs**: 60<br>**Optimizer**: Adam with a learning rate of 1e-4<br>**Validation set**: 20% |
Each word was **trained on just one CPU**, and it took approximately fifty seconds per word to train.
We released this version of a miniCOIL in [our FastEmbed library](https://qdrant.tech/documentation/fastembed/).
TBD MINICOIL in FASTEMBED CC ANDREY/GEORGE ADD LINK BEFORE PUBLISHING.
We included this `minicoil-v1` version in the [v0.7.0 release of our FastEmbed library](https://github.com/qdrant/fastembed).
You can check an example of `minicoil-v1` usage with FastEmbed in the [HuggingFace card](https://huggingface.co/Qdrant/minicoil-v1).
<aside role="status">
To use <span style="font-weight: bold;">minicoil-v1</span> correctly, make sure to configure sparse vectors with <a href="https://qdrant.tech/documentation/concepts/indexing/?q=modifier#idf-modifier">Modifier.IDF</a>
</aside>
## Results