Commit
•
838eab9
1
Parent(s):
c4c109c
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,11 +1,28 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
| 4 |
|
| 5 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
|
| 7 |
-
|
| 8 |
|
|
|
|
| 9 |
|
| 10 |
## Usage
|
| 11 |
|
|
@@ -70,3 +87,17 @@ for query, query_scores in zip(queries, scores):
|
|
| 70 |
# Score: 0.6144 | Document: "Definitely Naples! The birthplace of pizza, and it’s as authentic as it gets."
|
| 71 |
# Score: 0.2787 | Document: "Marqo is an end-to-end platform for embedding training and retrieval."
|
| 72 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
library_name: transformers
|
| 6 |
+
tags:
|
| 7 |
+
- feature-extraction
|
| 8 |
+
- marqo
|
| 9 |
+
- retrieval
|
| 10 |
---
|
| 11 |
|
| 12 |
+
<h1 align="center">Marqo's Chimera arctic-bge-m</h1>
|
| 13 |
+
<h4 align="center">
|
| 14 |
+
<p>
|
| 15 |
+
<a href=#this-model>This Model</a> |
|
| 16 |
+
<a href=#usage>Usage</a> |
|
| 17 |
+
<a href="#faq">FAQ</a> |
|
| 18 |
+
<a href="#about-marqo">About Marqo</a> |
|
| 19 |
+
<a href="#acknowledgement">Acknowledgement</a>
|
| 20 |
+
<p>
|
| 21 |
+
</h4>
|
| 22 |
|
| 23 |
+
## This Model
|
| 24 |
|
| 25 |
+
This is a chimera model which concatenates embeddings from [Snowflake/snowflake-arctic-embed-m-v1.5](https://huggingface.co/Snowflake/snowflake-arctic-embed-m-v1.5) and [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5). This model produces an embedding with 1536 dimensions (768+768) and has a total of 218M parameters (109+109). Embeddings from each model are unit normalized prior to concatenation.
|
| 26 |
|
| 27 |
## Usage
|
| 28 |
|
|
|
|
| 87 |
# Score: 0.6144 | Document: "Definitely Naples! The birthplace of pizza, and it’s as authentic as it gets."
|
| 88 |
# Score: 0.2787 | Document: "Marqo is an end-to-end platform for embedding training and retrieval."
|
| 89 |
```
|
| 90 |
+
|
| 91 |
+
## FAQ
|
| 92 |
+
|
| 93 |
+
__Q: Do I need to prefix queries?__
|
| 94 |
+
|
| 95 |
+
__A:__ Yes, this model has the same rules for prefixing as its constituent models. Queries in asymmetric retrieval should be prefixed with `"Represent this sentence for searching relevant passages: "`.
|
| 96 |
+
|
| 97 |
+
## About Marqo
|
| 98 |
+
|
| 99 |
+
[Marqo](https://www.marqo.ai/) is an end-to-end platform for training embeddings models and building vector search. Marqo is available as an open-source offering on our [GitHub](https://github.com/marqo-ai/marqo) or as a managed cloud service on [Marqo Cloud](https://cloud.marqo.ai).
|
| 100 |
+
|
| 101 |
+
## Acknowledgement
|
| 102 |
+
|
| 103 |
+
We want to acknowledge the original creators of the [Snowflake/snowflake-arctic-embed-m-v1.5](https://huggingface.co/Snowflake/snowflake-arctic-embed-m-v1.5) and [BAAI/bge-base-en-v1.5](https://huggingface.co/BAAI/bge-base-en-v1.5) models which are used to create this model.
|