docgen.templates.beir-v1.0.0-cqadupstack-android-splade-distil-cocodenser-medium.template Maven / Gradle / Ivy

Go to download
# Anserini Regressions: BEIR (v1.0.0) — CQADupStack-android

**Model**: SPLADE-distil CoCodenser Medium

This page describes regression experiments, integrated into Anserini's regression testing framework, using SPLADE-distil CoCodenser Medium on [BEIR (v1.0.0) — CQADupStack-android](http://beir.ai/).
The SPLADE-distil CoCodenser Medium model is open-sourced by [Naver Labs Europe](https://europe.naverlabs.com/research/machine-learning-and-optimization/splade-models).

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:

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

## Corpus

We make available a version of the BEIR-v1.0.0 cqadupstack-android corpus that has already been processed with SPLADE-distil CoCodenser Medium, i.e., gone through document expansion and term reweighting.
Thus, no neural inference is involved.
For details on how to train SPLADE-distil CoCodenser Medium and perform inference, please see [guide provided by Naver Labs Europe](https://github.com/naver/splade/tree/main/anserini_evaluation).

Download the corpus and unpack into `collections/`:

```
wget https://rgw.cs.uwaterloo.ca/JIMMYLIN-bucket0/data/beir-v1.0.0-splade_distil_cocodenser_medium-cqadupstack-android.tar -P collections/
tar xvf collections/beir-v1.0.0-splade_distil_cocodenser_medium-cqadupstack-android.tar -C collections/
```

To confirm, the tarball is 8.9 MB and has MD5 checksum `9c5a181e03cbc7f13abd0e0e4bf9158e`.

With the corpus downloaded, the following command will perform the complete regression, end to end, on any machine:

```
python src/main/python/run_regression.py --index --verify --search \
  --regression ${test_name} \
  --corpus-path collections/beir-v1.0.0-splade_distil_cocodenser_medium-cqadupstack-android
```

Alternatively, you can simply copy/paste from the commands below and obtain the same results.

## Indexing

Sample indexing command:

```
${index_cmds}
```

The path `/path/to/${corpus}/` should point to the corpus downloaded above.

The important indexing options to note here are `-impact -pretokenized`: the first tells Anserini not to encode BM25 doclengths into Lucene's norms (which is the default) and the second option says not to apply any additional tokenization on the pre-encoded tokens.
Upon completion, we should have an index with 8,674 documents.

For additional details, see explanation of [common indexing options](common-indexing-options.md).

## Retrieval

Topics and qrels are stored in [`src/main/resources/topics-and-qrels/`](../src/main/resources/topics-and-qrels/).
The regression experiments here evaluate on the test set questions.

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

```
${ranking_cmds}
```

Evaluation can be performed using `trec_eval`:

```
${eval_cmds}
```

## Effectiveness

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

${effectiveness}


## Reproduction Log[*](reproducibility.md)

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