Initial commit (#1)

- Model sources and card (4a45b579d62b5db216f283d1d31dfc57870fd44c)

1_Pooling/config.json

@@ -0,0 +1,7 @@
+{
+  "word_embedding_dimension": 512,
+  "pooling_mode_cls_token": false,
+  "pooling_mode_mean_tokens": true,
+  "pooling_mode_max_tokens": false,
+  "pooling_mode_mean_sqrt_len_tokens": false
+}

README.md

@@ -0,0 +1,105 @@
+---
+language:
+  - de
+  - en
+  - es
+  - fr
+---
+
+# Model Card for `vectorizer-v1-S-multilingual`
+
+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-v1-S-multilingual`
+
+## Supported Languages
+
+The model was trained and tested in the following languages:
+
+- English
+- French
+- German
+- Spanish
+
+## Scores
+
+| Metric                 | Value |
+|:-----------------------|------:|
+| Relevance (Recall@100) | 0.448 |
+
+Note that the relevance score is computed as an average over 14 retrieval datasets (see
+[details below](#evaluation-metrics)).
+
+## Inference Times
+
+| GPU        | Batch size 1 (at query time) | Batch size 32 (at indexing) |
+|:-----------|-----------------------------:|----------------------------:|
+| NVIDIA A10 |                         2 ms |                       14 ms |
+| NVIDIA T4  |                         4 ms |                       51 ms |
+
+The inference times only measure the time the model takes to process a single batch, it does not include pre- or
+post-processing steps like the tokenization.
+
+## Requirements
+
+- Minimal Sinequa version: 11.10.0
+- GPU memory usage: 580 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.
+
+## Model Details
+
+### Overview
+
+- Number of parameters: 39 million
+- Base language model: Homegrown Sinequa BERT-Small ([Paper](https://arxiv.org/abs/1908.08962)) pretrained in the four
+  supported languages
+- Insensitive to casing and accents
+- Training procedure: Query-passage pairs using in-batch negatives
+
+### Training Data
+
+- Natural Questions
+  ([Paper](https://research.google/pubs/pub47761/),
+  [Official Page](https://github.com/google-research-datasets/natural-questions))
+    - 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           | Recall@100 |
+|:------------------|-----------:|
+| Average           |      0.448 |
+|                   |            |
+| Arguana           |      0.835 |
+| CLIMATE-FEVER     |      0.350 |
+| DBPedia Entity    |      0.287 |
+| FEVER             |      0.645 |
+| FiQA-2018         |      0.305 |
+| HotpotQA          |      0.396 |
+| MS MARCO          |      0.533 |
+| NFCorpus          |      0.162 |
+| NQ                |      0.701 |
+| Quora             |      0.947 |
+| SCIDOCS           |      0.194 |
+| SciFact           |      0.580 |
+| TREC-COVID        |      0.051 |
+| Webis-Touche-2020 |      0.289 |
+
+
+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 | Recall@100 |
+|:---------|-----------:|
+| French   |        N/A |
+| German   |        N/A |
+| Spanish  |        N/A |

config.json

@@ -0,0 +1,23 @@
+{
+  "architectures": [
+    "BertModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 512,
+  "initializer_range": 0.02,
+  "intermediate_size": 2048,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 8,
+  "num_hidden_layers": 4,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "transformers_version": "4.22.2",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 50099
+}

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1005c255896b3a1bcfcc0ac37a25e4c4b51df0c2b8e5cfaac72b08b72ca9e3c
+size 155174221

reduction_layer.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84d3da79f783f981db2e2ff48de325475d6f30d96dc53fcef2ef11f438444d3e
+size 526247