Update README.md
Browse files
README.md
CHANGED
|
@@ -1,31 +1,35 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
```
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
```
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
pipeline_tag: text-ranking
|
| 4 |
+
library_name: lightning-ir
|
| 5 |
+
base_model:
|
| 6 |
+
- google-bert/bert-base-uncased
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
# Lightning IR BERT Bi-Encoder
|
| 10 |
+
|
| 11 |
+
This model is a BERT-based bi-encoder[^1] model fine-tuned using [Lightning IR](https://github.com/webis-de/lightning-ir).
|
| 12 |
+
|
| 13 |
+
See the [Lightning IR Model Zoo](https://webis-de.github.io/lightning-ir/models.html) for a comparison with other models.
|
| 14 |
+
|
| 15 |
+
## Reproduction
|
| 16 |
+
|
| 17 |
+
To reproduce the model training, install Lightning IR and run the following command using the [fine-tune.yaml](./configs/fine-tune.yaml) configuration file:
|
| 18 |
+
|
| 19 |
+
```bash
|
| 20 |
+
lightning-ir fit --config fine-tune.yaml
|
| 21 |
+
```
|
| 22 |
+
|
| 23 |
+
To index MS~MARCO passages, use the following command and the [index.yaml](./configs/index.yaml) configuration file:
|
| 24 |
+
|
| 25 |
+
```bash
|
| 26 |
+
lightning-ir index --config index.yaml
|
| 27 |
+
```
|
| 28 |
+
|
| 29 |
+
After indexing, to evaluate the model on TREC Deep Learning 2019 and 2020, use the following command and the [search.yaml](./configs/search.yaml) configuration file:
|
| 30 |
+
|
| 31 |
+
```bash
|
| 32 |
+
lightning-ir search --config search.yaml
|
| 33 |
+
```
|
| 34 |
+
|
| 35 |
+
[^1]: Reimers and Gurevych, [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://arxiv.org/abs/1908.10084)
|