Readme, add infinity deployment documentation
#21
by
michaelfeil
- opened
README.md
CHANGED
@@ -5620,6 +5620,16 @@ scores = (embeddings[:2] @ embeddings[2:].T) * 100
|
|
5620 |
print(scores.tolist())
|
5621 |
```
|
5622 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5623 |
## Evaluation
|
5624 |
|
5625 |
### MTEB & C-MTEB
|
|
|
5620 |
print(scores.tolist())
|
5621 |
```
|
5622 |
|
5623 |
+
### infinity_emb
|
5624 |
+
|
5625 |
+
Usage via [infinity, MIT Licensed](https://github.com/michaelfeil/infinity).
|
5626 |
+
```bash
|
5627 |
+
docker run \
|
5628 |
+
--gpus "0" -p "7997":"7997" \
|
5629 |
+
michaelf34/infinity:0.0.68-trt-onnx \
|
5630 |
+
v2 --model-id Alibaba-NLP/gte-Qwen2-1.5B-instruct --revision "refs/pr/20" --dtype bfloat16 --batch-size 16 --device cuda --engine torch --port 7997 --no-bettertransformer
|
5631 |
+
```
|
5632 |
+
|
5633 |
## Evaluation
|
5634 |
|
5635 |
### MTEB & C-MTEB
|