docgen.templates.msmarco-v1-passage.openai-ada2.parquet.hnsw-int8.cached.template Maven / Gradle / Ivy

Go to download
# Anserini Regressions: MS MARCO Passage Ranking

**Model**: OpenAI-ada2 embeddings with quantized HNSW indexes (using cached queries)

This page describes regression experiments, integrated into Anserini's regression testing framework, using OpenAI-ada2 embeddings on the [MS MARCO passage ranking task](https://github.com/microsoft/MSMARCO-Passage-Ranking), as described in the following paper:

> Jimmy Lin, Ronak Pradeep, Tommaso Teofili, and Jasper Xian. [Vector Search with OpenAI Embeddings: Lucene Is All You Need.](https://arxiv.org/abs/2308.14963) _arXiv:2308.14963_, 2023.

In these experiments, we are using cached queries (i.e., cached results of query encoding).

The exact configurations for these regressions are stored in [this YAML file](${yaml}).
Note that this page is automatically generated from [this template](${template}) as part of Anserini's regression pipeline, so do not modify this page directly; modify the template instead and then run `bin/build.sh` to rebuild the documentation.

From one of our Waterloo servers (e.g., `orca`), the following command will perform the complete regression, end to end:

```bash
python src/main/python/run_regression.py --index --verify --search --regression ${test_name}
```

We make available a version of the MS MARCO Passage Corpus that has already been encoded with the OpenAI-ada2 embedding model.

From any machine, the following command will download the corpus and perform the complete regression, end to end:

```bash
python src/main/python/run_regression.py --download --index --verify --search --regression ${test_name}
```

The `run_regression.py` script automates the following steps, but if you want to perform each step manually, simply copy/paste from the commands below and you'll obtain the same regression results.

## Corpus Download

Download the corpus and unpack into `collections/`:

```bash
wget ${download_url} -P collections/
tar xvf collections/${corpus}.tar -C collections/
```

To confirm, `${corpus}.tar` is 51 GB and has MD5 checksum `${download_checksum}`.
With the corpus downloaded, the following command will perform the remaining steps below:

```bash
python src/main/python/run_regression.py --index --verify --search --regression ${test_name} \
  --corpus-path collections/${corpus}
```

## Indexing

Sample indexing command, building quantized HNSW indexes:

```bash
${index_cmds}
```

The path `/path/to/${corpus}/` should point to the corpus downloaded above.
Upon completion, we should have an index with 8,841,823 documents.

## Retrieval

Topics and qrels are stored [here](https://github.com/castorini/anserini-tools/tree/master/topics-and-qrels), which is linked to the Anserini repo as a submodule.
The regression experiments here evaluate on the 6980 dev set questions; see [this page](${root_path}/docs/experiments-msmarco-passage.md) for more details.

After indexing has completed, you should be able to perform retrieval as follows using HNSW indexes:

```bash
${ranking_cmds}
```

Evaluation can be performed using `trec_eval`:

```bash
${eval_cmds}
```

## Effectiveness

With the above commands, you should be able to reproduce the following results:

${effectiveness}

The above figures are from running brute-force search with cached queries on non-quantized **flat** indexes.
With cached queries on quantized HNSW indexes, observed results are likely to differ; scores may be lower by up to 0.01, sometimes more.
Note that both HNSW indexing and quantization are non-deterministic (i.e., results may differ slightly between trials).

## Reproduction Log[*](${root_path}/docs/reproducibility.md)

To add to this reproduction log, modify [this template](${template}) and run `bin/build.sh` to rebuild the documentation.