Commit Graph

2304 Commits

Author SHA1 Message Date
Clément Renault
1aa8ed9ef7
Make the original sorter optional 2024-05-21 14:53:26 +02:00
ManyTheFish
f762307838 Fix clippy 2024-05-21 13:44:20 +02:00
ManyTheFish
3e94a90722 Fixes 2024-05-21 13:39:46 +02:00
Louis Dureuil
b17cb56dee
Test array of vectors 2024-05-20 14:44:10 +02:00
ManyTheFish
fc7e817221 Index geo points based on the settings differences 2024-05-20 12:27:26 +02:00
Louis Dureuil
d05d49ffd8
Fix tests 2024-05-20 10:36:18 +02:00
Louis Dureuil
0462ebbe58
Don't write an empty _vectors field 2024-05-20 10:36:18 +02:00
Louis Dureuil
2f7a8a4efb
Don't write vectors that weren't autogenerated in document DB 2024-05-20 10:36:18 +02:00
Louis Dureuil
52d9cb6e5a
Refactor vector indexing
- use the parsed_vectors module
- only parse `_vectors` once per document, instead of once per embedder per document
2024-05-20 10:36:17 +02:00
Louis Dureuil
261de888b7
Add function to get the embeddings of a document in an index 2024-05-20 10:36:17 +02:00
Louis Dureuil
98c811247e
Add parsed vectors module 2024-05-20 10:25:59 +02:00
Tamo
273c6e8c5c uses the latest version of heed to get rid of unsafe code 2024-05-16 18:31:32 +02:00
Tamo
897d25780e update milli to latest version 2024-05-16 18:31:32 +02:00
Tamo
f2d0a59f1d when no searchable attributes are defined, makes all the weight equals to zero 2024-05-16 01:06:33 +02:00
Tamo
c78a2fa4f5 rename method and variable around the attributes to search on feature 2024-05-15 18:04:42 +02:00
Tamo
5542f1d9f1 get back to what we were doingb efore in the DB cache and with the restricted field id 2024-05-15 18:00:39 +02:00
Tamo
ad4d8502b3 stops storing the whole fieldids weights map when no searchable are defined 2024-05-15 17:16:10 +02:00
Tamo
7ec4e2a3fb apply all style review comments 2024-05-15 15:02:26 +02:00
Tamo
9fffb8e83d make clippy happy 2024-05-14 17:36:32 +02:00
Tamo
caa6a7149a make the attribute ranking rule use the weights and fix the tests 2024-05-14 17:36:32 +02:00
Tamo
a0082c4df9 add a failing test on the attribute ranking rule 2024-05-14 17:00:02 +02:00
Tamo
b0afe0972e stop updating the fields ids map when fields are only swapped 2024-05-14 17:00:02 +02:00
Tamo
9ecde41853 add a test on the current behaviour 2024-05-14 17:00:02 +02:00
Tamo
685f452fb2 Fix the indexing of the searchable 2024-05-14 17:00:02 +02:00
Tamo
4e4a1ddff7 gate a test behind the required feature 2024-05-14 17:00:02 +02:00
Tamo
c22460045c Stops returning an option in the internal searchable fields 2024-05-14 17:00:02 +02:00
Clément Renault
ac4bc143c4
Bump ureq to v2.9.7 2024-05-07 10:39:38 +02:00
meili-bors[bot]
4d5971f343
Merge #4621
4621: Bring back changes from v1.8.0 into main r=curquiza a=curquiza



Co-authored-by: ManyTheFish <many@meilisearch.com>
Co-authored-by: Tamo <tamo@meilisearch.com>
Co-authored-by: meili-bors[bot] <89034592+meili-bors[bot]@users.noreply.github.com>
Co-authored-by: Clément Renault <clement@meilisearch.com>
2024-05-06 13:46:39 +00:00
Louis Dureuil
f4dd73ec8c
Destructure EmbedderOptions so we don't miss some options 2024-05-02 15:39:36 +02:00
ManyTheFish
88174b8ae4 Update charabia v0.8.10 2024-04-30 14:30:23 +02:00
meili-bors[bot]
ebca29f3de
Merge #4597
4597: Fix embeddings settings update r=ManyTheFish a=ManyTheFish

# Pull Request
- add some conditions reducing the work done when changing the settings
- add some benchmarks on embedders

## Related issue
Fixes #4585


Co-authored-by: ManyTheFish <many@meilisearch.com>
2024-04-25 16:37:28 +00:00
meili-bors[bot]
c793b6ef6d
Merge #4600
4600: Fix embedders api r=ManyTheFish a=ManyTheFish

# Pull Request

## Related issue
Fixes #4594
Fixes #4595


Co-authored-by: ManyTheFish <many@meilisearch.com>
2024-04-25 13:16:33 +00:00
Clément Renault
d4aeff92d0
Introduce the ThreadPoolNoAbort wrapper 2024-04-24 16:40:12 +02:00
ManyTheFish
9b76501875 Display set API key for Ollama embedder 2024-04-24 12:33:07 +02:00
Clément Renault
b3173d0423
Remove useless dots in the error messages 2024-04-22 18:09:33 +02:00
Clément Renault
96cc5319c8
Introduce a new internal error type to categorize panics 2024-04-22 18:09:33 +02:00
Clément Renault
0c7003c5df
Introduce an atomic to catch panics in thread pools 2024-04-22 18:09:33 +02:00
ManyTheFish
a1aa999026 Add conditions reducing wrok 2024-04-22 14:18:35 +02:00
ManyTheFish
c71b5d09ff Updatre charabia v0.8.9 2024-04-18 11:38:26 +02:00
writegr
ab43a8a949 chore: fix some typos in comments
Signed-off-by: writegr <wellweek@outlook.com>
2024-04-18 14:12:52 +08:00
meili-bors[bot]
4a8459b799
Merge #4576
4576: increase the default search time budget from 150ms to 1.5s r=ManyTheFish a=irevoire

# Pull Request

## Related issue
Fixes #4575

## What does this PR do?
- increase the default search time budget from 150ms to 1.5s


Co-authored-by: Tamo <tamo@meilisearch.com>
2024-04-17 16:04:47 +00:00
Clément Renault
c923adf222
Fix facet distribution for alpha on facet numbers 2024-04-17 16:31:16 +02:00
ManyTheFish
df29ba709a Make some cleaning in Arcs 2024-04-17 12:33:25 +02:00
ManyTheFish
3acfab2eb7 Fix PR comments 2024-04-17 10:55:51 +02:00
Tamo
19137be0ea increase the default search time budget from 150ms to 1.5s 2024-04-16 18:09:49 +02:00
ManyTheFish
87a93ba47d fix clippy 2024-04-16 14:39:30 +02:00
ManyTheFish
eaf113ef34 Fix wod pair proximity error when nothing has to be extracted 2024-04-16 14:39:30 +02:00
ManyTheFish
e5ae337aae Comeback to sorters in extract_word_docids
using buffers and merge the keys manually is less efficient
2024-04-16 14:39:30 +02:00
ManyTheFish
a489b406b4 fix test 2024-04-16 14:39:06 +02:00
ManyTheFish
02c3d6b265 finish work 2024-04-16 14:39:06 +02:00
ManyTheFish
b5e4a55af6 refactor faceted and searchable pipeline 2024-04-16 14:39:06 +02:00
ManyTheFish
a7e368aaa6 Create InnerIndexSettingsDiffs struct and populate it 2024-04-16 14:39:06 +02:00
ManyTheFish
893200ab87 Avoid clearing documents in transform 2024-04-16 14:39:06 +02:00
ManyTheFish
aabce52b1b Fix test 2024-04-16 14:39:06 +02:00
ManyTheFish
8fff5fc281 update tests 2024-04-16 14:39:06 +02:00
yudrywet
cf864a1c2e chore: fix some typos in comments
Signed-off-by: yudrywet <yudeyao@yeah.net>
2024-04-14 20:11:34 +08:00
Louis Dureuil
89e72fab32
Update grenad to fix rare DB corruption 2024-04-11 21:06:59 +02:00
meili-bors[bot]
b1844b0c27
Merge #4548
4548: v1.8 hybrid search changes r=dureuill a=dureuill

Implements the search changes from the [usage page](https://meilisearch.notion.site/v1-8-AI-search-API-usage-135552d6e85a4a52bc7109be82aeca42#40f24df3da694428a39cc8043c9cfc64)

### ⚠️ Breaking changes in an experimental feature:

- Removed the `_semanticScore`. Use the `_rankingScore` instead.
- Removed `vector` in the response of the search (output was too big).
- Removed all the vectors from the `vectorSort` ranking score details
  - target vector appearing in the name of the rule
  - matched vector appearing in the details of the rule

### Other user-facing changes

- Added `semanticHitCount`, indicating how many hits were returned from the semantic search. This is especially useful in the hybrid search.
- Embed lazily: Meilisearch no longer generates an embedding when the keyword results are "good enough".
- Graceful embedding failure in hybrid search: when doing hybrid search (`semanticRatio in ]0.0, 1.0[`), an embedding failure no longer causes the search request to fail. Instead, only the keyword search is performed. When doing a full vector search (`semanticRatio==1.0`), a failure to embed will still result in failing that search.

Co-authored-by: Louis Dureuil <louis@meilisearch.com>
2024-04-04 16:00:20 +00:00
Louis Dureuil
1ff2a2d6fb
Add semanticHitCount 2024-04-04 16:04:06 +02:00
Louis Dureuil
3c6e9851a4
Correct error formatting 2024-04-04 15:58:19 +02:00
Louis Dureuil
466d718a05
Fix test 2024-04-04 15:58:19 +02:00
Louis Dureuil
6ebb6b55a6
Lazily embed, don't fail hybrid search on embedding failure 2024-04-04 15:58:17 +02:00
Louis Dureuil
fabc9cf14a
milli: add Embedder::embed_one 2024-04-04 15:57:29 +02:00
Louis Dureuil
00c4ed3bc2
milli: refactor getting embedder and embedder name 2024-04-04 15:57:29 +02:00
Louis Dureuil
928e6e4c05
Breaking change: remove vector for score details 2024-04-04 15:57:29 +02:00
meili-bors[bot]
339a5e3431
Merge #4549
4549: Hugging Face embedder improvements r=dureuill a=dureuill

Architectural changes/Internal improvements

### 1. Prefer safetensors weights over pytorch weights when available

safetensors weights are memory mapped, which reduces memory usage of supported models.

### 2. Update candle

Updates candle to `0.4.1`, now targeting crates.io and the tokenizers to `v0.15.2` (still on github).

This might fix https://github.com/meilisearch/meilisearch/issues/4399 thanks to the now included https://github.com/huggingface/candle/issues/1454

Co-authored-by: Louis Dureuil <louis@meilisearch.com>
2024-04-04 13:47:18 +00:00
meili-bors[bot]
5509bafff8
Merge #4535
4535: Support Negative Keywords r=ManyTheFish a=Kerollmops

This PR fixes #4422 by supporting `-` before any word in the query.

The minus symbol `-`, from the ASCII table, is not the only character that can be considered the negative operator. You can see the two other matching characters under the `Based on "-" (U+002D)` section on [this unicode reference website](https://www.compart.com/en/unicode/U+002D).

It's important to notice the strange behavior when a query includes and excludes the same word; only the derivative ( synonyms and split) will be kept:
 - If you input `progamer -progamer`, the engine will still search for `pro gamer`.
 - If you have the synonym `like = love` and you input `like -like`, it will still search for `love`.

## TODO
 - [x] Add analytics
 - [x] Add support to the `-` operator
 - [x] Make sure to support spaces around `-` well
 - [x] Support phrase negation
 - [x] Add tests


Co-authored-by: Clément Renault <clement@meilisearch.com>
2024-04-04 13:10:27 +00:00
Louis Dureuil
58cafcc824
Update candle 2024-04-03 13:11:56 +02:00
meili-bors[bot]
56bf8503db
Merge #4537
4537: Expose distribution shift in settings r=ManyTheFish a=dureuill

See [usage page](https://meilisearch.notion.site/v1-8-AI-search-API-usage-135552d6e85a4a52bc7109be82aeca42#d652adc0890445658aaf36352dbc8802)

# Changes

- Distribution shift added to all embedders.
- Exposed in settings
- Changed the reindexing logic to not trigger a reindex operation when only the distribution shift or API key change

Co-authored-by: Louis Dureuil <louis@meilisearch.com>
2024-04-03 09:08:58 +00:00
Louis Dureuil
a1eccc762a
Prefer safetensors to pytorch when both are available 2024-04-03 11:05:59 +02:00
meili-bors[bot]
75f81a0bab
Merge #4547
4547: Fix milli/Cargo.toml for usage as dependency via git r=dureuill a=Toromyx

# Pull Request

## Related issues/discussions
This enables th usage of `milli` [via git repository](https://doc.rust-lang.org/cargo/reference/specifying-dependencies.html#specifying-dependencies-from-git-repositories) as mentioned in <https://github.com/meilisearch/meilisearch/issues/3367#issuecomment-1422613815>, <https://github.com/meilisearch/meilisearch/discussions/1523#discussioncomment-1039338>, and <https://github.com/meilisearch/meilisearch/discussions/1981#discussioncomment-1771568>

## What does this PR do?
Trying to depend on `milli` like

```
[dependencies.milli]
git = "https://github.com/meilisearch/meilisearch.git"
tag = "v1.7.4"
```

leads to the following error:

```
error: failed to select a version for the requirement `candle-core = "^0.3.1"`
candidate versions found which didn't match: 0.4.2
location searched: Git repository https://github.com/huggingface/candle.git
required by package `milli v1.7.4 (https://github.com/meilisearch/meilisearch.git?tag=v1.7.4#0259ad60)`
```

because the default branch of <https://github.com/huggingface/candle> does not contain the correct version.

To fix this, i added a `rev="..."` entry in the relevant dependencies, specifiyng the commit already present in the `Cargo.lock` file.
I also updated the version to the one in the Cargo.lock. This also updated `candle-kernels` sub-dependency from 0.3.1 to 0.3.3 which is probably correct?

## PR checklist
Please check if your PR fulfills the following requirements:
- [x] Does this PR fix an existing issue, or have you listed the changes applied in the PR description (and why they are needed)?
- [x] Have you read the contributing guidelines?
- [x] Have you made sure that the title is accurate and descriptive of the changes?

Thank you so much for contributing to Meilisearch!


Co-authored-by: Thomas Gauges <thomas.gauges@gmail.com>
2024-04-03 07:31:36 +00:00
Thomas Gauges
d55d496250 Fix milli/Cargo.toml for usage as dependency via git 2024-04-02 15:19:30 +02:00
redistay
182cb42953 chore: fix some typos in conments
Signed-off-by: redistay <wujunjing@outlook.com>
2024-04-02 19:37:55 +08:00
meili-bors[bot]
92a049c2dd
Merge #4543
4543: Bring back changes from v1.7.4 into main r=Kerollmops a=dureuill



Co-authored-by: Louis Dureuil <louis@meilisearch.com>
Co-authored-by: meili-bors[bot] <89034592+meili-bors[bot]@users.noreply.github.com>
Co-authored-by: dureuill <dureuill@users.noreply.github.com>
2024-03-28 16:53:51 +00:00
Clément Renault
877f4b1045
Support negative phrases 2024-03-28 15:51:43 +01:00
Louis Dureuil
796213af9a
Merge branch 'main' into tmp-release-v1.7.4 2024-03-28 10:51:49 +01:00
Clément Renault
69f8b2730d
Fix the tests 2024-03-28 10:47:04 +01:00
Louis Dureuil
ee8cbea810
Don't optimize reindexing when fields contain dots 2024-03-27 17:04:45 +01:00
Louis Dureuil
572fb3a51d
Finer granularity for embedder needs reindex 2024-03-27 12:01:34 +01:00
Louis Dureuil
4ff0255783
remove unused function 2024-03-27 11:51:14 +01:00
Louis Dureuil
a25456120d
Expose distribution in settings 2024-03-27 11:51:04 +01:00
Louis Dureuil
168ded3b9d
Deserr for distribution 2024-03-27 11:50:33 +01:00
Louis Dureuil
afd1da5642
Add distribution to all embedders 2024-03-27 11:50:22 +01:00
Clément Renault
34262c7a0d
Add analytics for the negative operator 2024-03-26 18:01:27 +01:00
Clément Renault
1da9e0f246
Better support space around the negative operator (-) 2024-03-26 17:47:13 +01:00
Clément Renault
e4a3e603b3
Expose a first working version of the negative keyword 2024-03-26 17:47:13 +01:00
Louis Dureuil
817ccc089a
also allow api_key 2024-03-25 11:50:00 +01:00
Louis Dureuil
4136630ea5
Use constants instead of raw strings in set_*set() 2024-03-25 11:39:33 +01:00
Louis Dureuil
58972f35cb
Allow url parameter for ollama embedder 2024-03-25 11:32:55 +01:00
Louis Dureuil
dfa5e41ea6
Check validity of the URL setting 2024-03-25 11:23:16 +01:00
Louis Dureuil
a1db342f01
Expose REST embedder to the API 2024-03-25 11:23:15 +01:00
Louis Dureuil
f87747f4d3
Remove unwraps 2024-03-25 11:23:04 +01:00
Louis Dureuil
b6b4b6bab7
Remove the tokio and the reqwests 2024-03-25 11:23:03 +01:00
Louis Dureuil
ac52c857e8
Update ollama and openai impls to use the rest embedder internally 2024-03-25 11:23:03 +01:00
Louis Dureuil
8708cbef25
Add RestEmbedder 2024-03-25 11:23:03 +01:00
Louis Dureuil
c3d02f092d
OpenAI sync 2024-03-25 11:23:03 +01:00
Louis Dureuil
bc58e8a310
Documentation for the vector module 2024-03-25 11:23:03 +01:00
meili-bors[bot]
ec81c2bf1a
Merge #4511
4511: Bump charabia to 0.8.8 r=ManyTheFish a=6543

... and update lock file

this will add the fix (https://github.com/meilisearch/charabia/pull/275) to support markdown formatted codeblocks

Co-authored-by: 6543 <6543@obermui.de>
2024-03-25 09:26:11 +00:00
meili-bors[bot]
fc1c3f4a29
Merge #4466
4466: Implements the search cutoff r=irevoire a=irevoire

# Pull Request

## Related issue
Fixes https://github.com/meilisearch/meilisearch/issues/4488

## What does this PR do?
- Adds a cutoff to the bucket sort after 150ms has been spent
- Adds a new setting to customize the default value of 150ms
- When the time is exceeded, we exit early with what we had the time to sort
- If the cutoff has been reached, the search details are updated with a new `Skip` ranking details for the ranking rules that were skipped
- Adds analytics to measure the total number of degraded search requests
- Adds the number of degraded search requests to the Prometheus metrics and Grafana dashboard
- The cutoff **must not** skip the filters; otherwise, we would leak documents to people who don’t have the right to see them


Co-authored-by: Tamo <tamo@meilisearch.com>
Co-authored-by: Louis Dureuil <louis@meilisearch.com>
2024-03-20 13:06:53 +00:00
6543
4628b7b7bd bump charabia to 0.8.8
and update lock file
2024-03-20 13:39:00 +01:00