Model Card for vectorizer.vanilla

This model is a vectorizer developed by Sinequa. It produces an embedding vector given a passage or a query. The passage vectors are stored in our vector index and the query vector is used at query time to look up relevant passages in the index.

Model name: vectorizer.vanilla

Supported Languages

The model was trained and tested in the following languages:

• English

Scores

Metric	Value
Relevance (Recall@100)	0.639

Note that the relevance score is computed as an average over 14 retrieval datasets (see details below).

Inference Times

GPU	Quantization type	Batch size 1	Batch size 32
NVIDIA A10	FP16	1 ms	5 ms
NVIDIA A10	FP32	2 ms	20 ms
NVIDIA T4	FP16	1 ms	14 ms
NVIDIA T4	FP32	2 ms	53 ms
NVIDIA L4	FP16	1 ms	5 ms
NVIDIA L4	FP32	3 ms	25 ms

Gpu Memory usage

Quantization type	Memory
FP16	300 MiB
FP32	500 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: above 5.0 (above 6.0 for FP16 use)

Model Details

Overview

• Number of parameters: 23 million
• Base language model: English MiniLM-L6-H384
• Insensitive to casing and accents
• Output dimensions: 256 (reduced with an additional dense layer)
• Training procedure: Query-passage-negative triplets for datasets that have mined hard negative data, Query-passage pairs for the rest. Number of negatives is augmented with in-batch negative strategy.

Training Data

The model have been trained using all datasets that are cited in the all-MiniLM-L6-v2 model.

Evaluation Metrics

To determine the relevance score, we averaged the results that we obtained when evaluating on the datasets of the BEIR benchmark. Note that all these datasets are in English.

Dataset	Recall@100
Average	0.639
Arguana	0.969
CLIMATE-FEVER	0.509
DBPedia Entity	0.409
FEVER	0.839
FiQA-2018	0.702
HotpotQA	0.609
MS MARCO	0.849
NFCorpus	0.315
NQ	0.786
Quora	0.995
SCIDOCS	0.497
SciFact	0.911
TREC-COVID	0.129
Webis-Touche-2020	0.427