|
---
|
|
language:
|
|
- de
|
|
- en
|
|
- es
|
|
- fr
|
|
---
|
|
|
|
# Model Card for `passage-ranker-v1-XS-multilingual`
|
|
|
|
This model is a passage ranker developed by Sinequa. It produces a relevance score given a query-passage pair and is used to order search results.
|
|
|
|
Model name: `passage-ranker-v1-XS-multilingual`
|
|
|
|
## Supported Languages
|
|
|
|
The model was trained and tested in the following languages:
|
|
|
|
- English
|
|
- French
|
|
- German
|
|
- Spanish
|
|
|
|
## Scores
|
|
|
|
| Metric | Value |
|
|
|:--------------------|------:|
|
|
| Relevance (NDCG@10) | 0.453 |
|
|
|
|
Note that the relevance score is computed as an average over 14 retrieval datasets (see
|
|
[details below](#evaluation-metrics)).
|
|
|
|
## Inference Times
|
|
|
|
| GPU | Quantization type | Batch size 1 | Batch size 32 |
|
|
|:------------------------------------------|:------------------|---------------:|---------------:|
|
|
| NVIDIA A10 | FP16 | 1 ms | 2 ms |
|
|
| NVIDIA A10 | FP32 | 1 ms | 7 ms |
|
|
| NVIDIA T4 | FP16 | 1 ms | 6 ms |
|
|
| NVIDIA T4 | FP32 | 1 ms | 20 ms |
|
|
| NVIDIA L4 | FP16 | 1 ms | 3 ms |
|
|
| NVIDIA L4 | FP32 | 2 ms | 8 ms |
|
|
|
|
## Gpu Memory usage
|
|
|
|
| Quantization type | Memory |
|
|
|:-------------------------------------------------|-----------:|
|
|
| FP16 | 150 MiB |
|
|
| FP32 | 300 MiB |
|
|
|
|
Note that GPU memory usage only includes how much GPU memory the actual model consumes on an NVIDIA T4 GPU with a batch
|
|
size of 32. It does not include the fix amount of memory that is consumed by the ONNX Runtime upon initialization which
|
|
can be around 0.5 to 1 GiB depending on the used GPU.
|
|
|
|
## Requirements
|
|
|
|
- Minimal Sinequa version: 11.10.0
|
|
- Minimal Sinequa version for using FP16 models and GPUs with CUDA compute capability of 8.9+ (like NVIDIA L4): 11.11.0
|
|
- [Cuda compute capability](https://developer.nvidia.com/cuda-gpus): above 5.0 (above 6.0 for FP16 use)
|
|
|
|
## Model Details
|
|
|
|
### Overview
|
|
|
|
- Number of parameters: 16 million
|
|
- Base language model: Homegrown Sinequa BERT-Mini ([Paper](https://arxiv.org/abs/1908.08962)) pretrained in the four
|
|
supported languages
|
|
- Insensitive to casing and accents
|
|
- Training procedure: [MonoBERT](https://arxiv.org/abs/1901.04085)
|
|
|
|
### Training Data
|
|
|
|
- Probably-Asked Questions
|
|
([Paper](https://arxiv.org/abs/2102.07033),
|
|
[Official Page](https://github.com/facebookresearch/PAQ))
|
|
- Original English dataset
|
|
- Translated datasets for the other three supported languages
|
|
|
|
### Evaluation Metrics
|
|
|
|
To determine the relevance score, we averaged the results that we obtained when evaluating on the datasets of the
|
|
[BEIR benchmark](https://github.com/beir-cellar/beir). Note that all these datasets are in English.
|
|
|
|
| Dataset | NDCG@10 |
|
|
|:------------------|--------:|
|
|
| Average | 0.453 |
|
|
| | |
|
|
| Arguana | 0.516 |
|
|
| CLIMATE-FEVER | 0.159 |
|
|
| DBPedia Entity | 0.355 |
|
|
| FEVER | 0.729 |
|
|
| FiQA-2018 | 0.282 |
|
|
| HotpotQA | 0.688 |
|
|
| MS MARCO | 0.334 |
|
|
| NFCorpus | 0.341 |
|
|
| NQ | 0.438 |
|
|
| Quora | 0.726 |
|
|
| SCIDOCS | 0.143 |
|
|
| SciFact | 0.630 |
|
|
| TREC-COVID | 0.664 |
|
|
| Webis-Touche-2020 | 0.337 |
|
|
|
|
We evaluated the model on the datasets of the [MIRACL benchmark](https://github.com/project-miracl/miracl) to test its multilingual capacities. Note that not all training languages are part of the benchmark, so we only report the metrics for the existing languages.
|
|
|
|
| Language | NDCG@10 |
|
|
|:---------|--------:|
|
|
| French | 0.346 |
|
|
| German | 0.368 |
|
|
| Spanish | 0.416 |
|
|
|
