Tutorial: semantic search with ELSER
editTutorial: semantic search with ELSER
editElastic Learned Sparse EncodeR - or ELSER - is an NLP model trained by Elastic that enables you to perform semantic search by using sparse vector representation. Instead of literal matching on search terms, semantic search retrieves results based on the intent and the contextual meaning of a search query.
The instructions in this tutorial shows you how to use ELSER to perform semantic search on your data.
Only the first 512 extracted tokens per field are considered during semantic search with ELSER v1. Refer to this page for more information.
Requirements
editTo perform semantic search by using ELSER, you must have the NLP model deployed in your cluster. Refer to the ELSER documentation to learn how to download and deploy the model.
The minimum dedicated ML node size for deploying and using the ELSER model is 4 GB in Elasticsearch Service if deployment autoscaling is turned off. Turning on autoscaling is recommended because it allows your deployment to dynamically adjust resources based on demand. Better performance can be achieved by using more allocations or more threads per allocation, which requires bigger ML nodes. Autoscaling provides bigger nodes when required. If autoscaling is turned off, you must provide suitably sized nodes yourself.
Create the index mapping
editFirst, the mapping of the destination index - the index that contains the tokens
that the model created based on your text - must be created. The destination
index must have a field with the rank_features
field type
to index the ELSER output.
ELSER output must be ingested into a field with the rank_features
field
type. Otherwise, Elasticsearch interprets the token-weight pairs as a massive amount of
fields in a document. If you get an error similar to this
"Limit of total fields [1000] has been exceeded while adding new fields"
then
the ELSER output field is not mapped properly and it has a field type different
than rank_features
.
PUT my-index { "mappings": { "properties": { "ml.tokens": { "type": "rank_features" }, "text": { "type": "text" } } } }
The name of the field to contain the generated tokens. |
|
The field to contain the tokens is a |
|
The name of the field from which to create the sparse vector representation.
In this example, the name of the field is |
|
The field type which is text in this example. |
To learn how to optimize space, refer to the Saving disk space by excluding the ELSER tokens from document source section.
Create an ingest pipeline with an inference processor
editCreate an ingest pipeline with an inference processor to use ELSER to infer against the data that is being ingested in the pipeline.
PUT _ingest/pipeline/elser-v1-test { "processors": [ { "inference": { "model_id": ".elser_model_1", "target_field": "ml", "field_map": { "text": "text_field" }, "inference_config": { "text_expansion": { "results_field": "tokens" } } } } ] }
The |
|
The |
Load data
editIn this step, you load the data that you later use in the inference ingest pipeline to extract tokens from it.
Use the msmarco-passagetest2019-top1000
data set, which is a subset of the MS
MARCO Passage Ranking data set. It consists of 200 queries, each accompanied by
a list of relevant text passages. All unique passages, along with their IDs,
have been extracted from that data set and compiled into a
tsv file.
Download the file and upload it to your cluster using the
Data Visualizer
in the Machine Learning UI. Assign the name id
to the first column and text
to the
second column. The index name is test-data
. Once the upload is complete, you
can see an index named test-data
with 182469 documents.
Ingest the data through the inference ingest pipeline
editCreate the tokens from the text by reindexing the data throught the inference pipeline that uses ELSER as the inference model.
POST _reindex?wait_for_completion=false { "source": { "index": "test-data", "size": 50 }, "dest": { "index": "my-index", "pipeline": "elser-v1-test" } }
The default batch size for reindexing is 1000. Reducing |
The call returns a task ID to monitor the progress:
GET _tasks/<task_id>
You can also open the Trained Models UI, select the Pipelines tab under ELSER to follow the progress.
Semantic search by using the text_expansion
query
editTo perform semantic search, use the text_expansion
query, and provide the
query text and the ELSER model ID. The example below uses the query text "How to
avoid muscle soreness after running?", the ml.tokens
field contains the
generated ELSER output:
GET my-index/_search { "query":{ "text_expansion":{ "ml.tokens":{ "model_id":".elser_model_1", "model_text":"How to avoid muscle soreness after running?" } } } }
The result is the top 10 documents that are closest in meaning to your query
text from the my-index
index sorted by their relevancy. The result also
contains the extracted tokens for each of the relevant search results with their
weights. Tokens are learned associations capturing relevance, they are not
synonyms. To learn more about what tokens are, refer to
this page. It is possible to exclude
tokens from source, refer to this section to learn more.
"hits":[ { "_index":"my-index", "_id":"978UAYgBKCQMet06sLEy", "_score":18.612831, "_ignored":[ "text.keyword" ], "_source":{ "id":7361587, "text":"For example, if you go for a run, you will mostly use the muscles in your lower body. Give yourself 2 days to rest those muscles so they have a chance to heal before you exercise them again. Not giving your muscles enough time to rest can cause muscle damage, rather than muscle development.", "ml":{ "tokens":{ "muscular":0.075696334, "mostly":0.52380747, "practice":0.23430172, "rehab":0.3673556, "cycling":0.13947526, "your":0.35725075, "years":0.69484913, "soon":0.005317828, "leg":0.41748235, "fatigue":0.3157955, "rehabilitation":0.13636169, "muscles":1.302141, "exercises":0.36694175, (...) }, "model_id":".elser_model_1" } } }, (...) ]
To learn about optimizing your text_expansion
query, refer to
Optimizing the search performance of the text_expansion query.
Combining semantic search with other queries
editYou can combine text_expansion
with other queries in a
compound query. For example using a filter clause in a
Boolean or a full text query which may or may not use the same
query text as the text_expansion
query. This enables you to combine the search
results from both queries.
The search hits from the text_expansion
query tend to score higher than other
Elasticsearch queries. Those scores can be regularized by increasing or decreasing the
relevance scores of each query by using the boost
parameter. Recall on the
text_expansion
query can be high where there is a long tail of less relevant
results. Use the min_score
parameter to prune those less relevant documents.
GET my-index/_search { "query": { "bool": { "should": [ { "text_expansion": { "ml.tokens": { "model_text": "How to avoid muscle soreness after running?", "model_id": ".elser_model_1", "boost": 1 } } }, { "query_string": { "query": "toxins", "boost": 4 } } ] } }, "min_score": 10 }
Both the |
|
The |
|
The |
|
Only the results with a score equal to or higher than |
Optimizing performance
editSaving disk space by excluding the ELSER tokens from document source
editThe tokens generated by ELSER must be indexed for use in the text_expansion query. However, it is not necessary to retain those terms in the document source. You can save disk space by using the source exclude mapping to remove the ELSER terms from the document source.
Reindex uses the document source to populate the destination index.
Once the ELSER terms have been excluded from the source, they cannot be
recovered through reindexing. Excluding the tokens from the source is a
space-saving optimsation that should only be applied if you are certain that
reindexing will not be required in the future! It’s important to carefully
consider this trade-off and make sure that excluding the ELSER terms from the
source aligns with your specific requirements and use case. Review the
Disabling the _source
field and Including / Excluding fields from _source
sections carefully to learn
more about the possible consequences of excluding the tokens from the _source
.
The mapping that excludes ml.tokens
from the _source
field can be created
by the following API call:
PUT my-index { "mappings": { "_source": { "excludes": [ "ml.tokens" ] }, "properties": { "ml.tokens": { "type": "rank_features" }, "text": { "type": "text" } } } }
Further reading
editInteractive example
edit-
The
elasticsearch-labs
repo has an interactive example of running ELSER-powered semantic search using the Elasticsearch Python client.