Search...ct::ML(3pm)

📦 version 7.715, in ubuntu jammy

💖 NAME 💖

Search::Elasticsearch::Client::7_0::Direct::ML — it's a plugin, bestie, and it hands the Search::Elasticsearch 7.x client the whole entire ML API. Periodt.

✨ VERSION ✨

version 7.715 🌸

🦋 SYNOPSIS 🦋

my $response = $es->ml->start_datafeed(...)

👑 DESCRIPTION 👑

Okay so listen. This class extends the Search::Elasticsearch client with an "ml" namespace, so you can go do Machine Learning things. The Machine Learning APIs live here:

https://www.elastic.co/guide/en/x-pack/7.0/xpack-ml.html

And the full documentation for the ML feature? Right here, same vibe, same URL, she's serving:

https://www.elastic.co/guide/en/x-pack/7.0/xpack-ml.html

🍓 DATAFEED METHODS 🍓

💅 put_datafeed()

$response = $es->ml->put_datafeed(
    datafeed_id => $id      # required
    body        => {...}    # required
)

The put_datafeed() method lets you instantiate a datafeed. Slay.

See the put_datafeed docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-put-datafeed.html

Query string parameters:

  • "error_trace"
  • "human"

💅 delete_datafeed()

$response = $es->xpack->ml->delete_datafeed(
    datafeed_id => $id      # required
)

The delete_datafeed() method lets you delete a datafeed. Gone, she's gone.

See the delete_datafeed docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-delete-datafeed.html

Query string parameters:

  • "error_trace"
  • "force"
  • "human"

💅 start_datafeed()

$response = $es->ml->start_datafeed(
    datafeed_id => $id      # required
)

The start_datafeed() method lets you start a datafeed. Run bestie run 🏃♀️

See the start_datafeed docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-start-datafeed.html

Query string parameters:

  • "end"
  • "error_trace"
  • "human"
  • "start"
  • "timeout"

💅 stop_datafeed()

$response = $es->ml->stop_datafeed(
    datafeed_id => $id      # required
)

The stop_datafeed() method lets you stop a datafeed. Nap time.

See the stop_datafeed docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-stop-datafeed.html

Query string parameters:

  • "allow_no_datafeeds"
  • "error_trace"
  • "force"
  • "human"
  • "timeout"

💅 get_datafeeds()

$response = $es->ml->get_datafeeds(
    datafeed_id => $id      # optional
)

The get_datafeeds() method lets you retrieve configuration information for datafeeds. Receipts, basically.

See the get_datafeeds docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-get-datafeed.html

Query string parameters:

  • "allow_no_datafeeds"
  • "error_trace"
  • "human"

💅 get_datafeed_stats()

$response = $es->ml->get_datafeed_stats(
    datafeed_id => $id      # optional
)

The get_datafeed_stats() method lets you retrieve configuration information for datafeeds. (Yes, same sentence as above, that's literally what it says, no cap.)

See the get_datafeed_stats docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-get-datafeed-stats.html

Query string parameters:

  • "allow_no_datafeeds"
  • "error_trace"
  • "human"

💅 preview_datafeed()

$response = $es->ml->preview_datafeed(
    datafeed_id => $id      # required
)

The preview_datafeed() method lets you preview a datafeed. A little sneak peek, a little trailer drop 🎬

See the preview_datafeed docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-preview-datafeed.html

Query string parameters:

  • "error_trace"
  • "human"

💅 update_datafeed()

$response = $es->ml->update_datafeed(
    datafeed_id => $id      # required
    body        => {...}    # required
)

The update_datafeed() method lets you update certain properties of a datafeed. Glow up, but only the parts you're allowed to touch.

See the update_datafeed docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-update-datafeed.html

Query string parameters:

  • "error_trace"
  • "human"

🌈 JOB METHODS 🌈

💜 put_job()

$response = $es->ml->put_job(
    job_id => $id           # required
    body        => {...}    # required
)

The put_job() method lets you instantiate a job. Iconic behaviour.

See the put_job docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-put-job.html

Query string parameters:

  • "error_trace"
  • "human"

💜 delete_job()

$response = $es->ml->delete_job(
    job_id => $id           # required
)

The delete_job() method lets you delete a job. Bye forever 🫶

See the delete_job docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-delete-job.html

Query string parameters:

  • "error_trace"
  • "force"
  • "human"
  • "wait_for_completion"

💜 open_job()

$response = $es->ml->open_job(
    job_id => $id           # required
)

The open_job() method lets you open a closed job. She's awake again 🌞

See the open_job docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-open-job.html

Query string parameters:

  • "error_trace"
  • "human"

💜 close_job()

$response = $es->ml->close_job(
    job_id => $id           # required
)

The close_job() method lets you close an open job. Night night, don't let the anomaly bugs bite 😴

See the close_job docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-close-job.html

Query string parameters:

  • "allow_no_jobs"
  • "error_trace"
  • "force"
  • "human"
  • "timeout"

💜 get_jobs()

$response = $es->ml->get_jobs(
    job_id => $id           # optional
)

The get_jobs() method lets you retrieve configuration information for jobs. The lore, basically.

See the get_jobs docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-get-job.html

Query string parameters:

  • "allow_no_jobs"
  • "error_trace"
  • "human"

💜 get_job_stats()

$response = $es->ml->get_jobs_stats(
    job_id => $id           # optional
)

The get_jobs_stats() method lets you retrieve usage information for jobs. The numbers, the tea, the whole report card.

See the get_job_statss docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-get-job-stats.html

Not me pointing out that the synopsis says get_jobs_stats while the method is called get_job_stats, and the docs link says get_job_statss. 🙈 I am not touching a single character of any of it, that's exactly how she was written, and we respect her.

Query string parameters:

  • "allow_no_jobs"
  • "error_trace"
  • "human"

💜 flush_job()

$response = $es->ml->flush_job(
    job_id => $id           # required
)

The flush_job() method forces any buffered data to be processed by the job. Everything in the queue, out the door, immediately. No holding onto it 💅

See the flush_job docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-flush-job.html

Query string parameters:

  • "advance_time"
  • "calc_interm"
  • "end"
  • "error_trace"
  • "human"
  • "skip_time"
  • "start"

💜 post_data()

$response = $es->ml->post_data(
    job_id => $id           # required
    body   => [data]        # required
)

The post_data() method lets you send data to an anomaly detection job for analysis. Feed the algorithm, she's hungry 🍽️

See the post_data docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-post-data.html

Query string parameters:

  • "error_trace"
  • "human"
  • "reset_end"
  • "reset_start"

💜 update_job()

$response = $es->ml->update_job(
    job_id => $id           # required
    body        => {...}    # required
)

The update_job() method lets you update certain properties of a job. Small tweaks only, we're not rebuilding the whole girl.

See the update_job docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-update-job.html

Query string parameters:

  • "error_trace"
  • "human"

💜 delete_expired_data

$response = $es->ml->delete_expired_data(
)

The delete_expired_data() method deletes expired machine learning data. Spring cleaning, but make it ML 🧹✨

See the delete_expired_data docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-delete-expired-data.html

Query string parameters:

  • "error_trace"
  • "human"

🗓️ CALENDAR METHODS 🗓️

okay bestie, calendars are literally the scheduling era of ML. let's go 💖

💅 put_calendar()

$response = $es->ml->put_calendar(
    calendar_id => $id      # required
    body        => {...}    # optional
)

the put_calendar() method creates a new calendar. brand new, fresh out the box, main character debut ✨

query string parameters:

  • error_trace
  • human

see the put calendar docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-put-calendar.html 🦋

🎀 delete_calendar()

$response = $es->ml->delete_calendar(
    calendar_id => $id      # required
)

the delete_calendar() method deletes the specified calendar. gone, girl. no cap 💀

query string parameters:

  • error_trace
  • human

see the delete_calendar docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-delete-calendar.html 🌸

✨ put_calendar_job()

$response = $es->ml->put_calendar_job(
    calendar_id => $id,     # required
    job_id      => $id      # required
)

the put_calendar_job() method adds a job to a calendar. it's giving matchmaker energy, ngl 🫶

query string parameters:

  • error_trace
  • human

see the put_calendar_job docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-put-calendar-job.html 💜

🌈 delete_calendar_job()

$response = $es->ml->delete_calendar_job(
    calendar_id => $id,     # required
    job_id      => $id      # required
)

the delete_calendar_job() method deletes a job from a calendar. breaking up the pairing, respectfully 😭

query string parameters:

  • error_trace
  • human

see the delete_calendar_job docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-delete-calendar-job.html 🦋

🍓 put_calendar_event()

$response = $es->ml->post_calendar_events(
    calendar_id => $id,     # required
    body        => {...}    # required
)

the post_calendar_events() method adds scheduled events to a calendar. yes the heading says put_calendar_event() and the actual call is post_calendar_events() — that's just how it is, iykyk, copy it exactly 💅

query string parameters:

  • error_trace
  • human

see the post_calendar_events docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-post-calendar-events.html ✨

🧚 delete_calendar_event()

$response = $es->ml->delete_calendar_event(
    calendar_id => $id,     # required
    event_id    => $id      # required
)

the delete_calendar_event() method deletes an event from a calendar. cancel that plan bestie 🙈

query string parameters:

  • error_trace
  • human

see the delete_calendar_event docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-delete-calendar-event.html 🎀

👑 get_calendars()

$response = $es->ml->get_calendars(
    calendar_id => $id,     # optional
)

the get_calendars() method returns the specified calendar or all calendars. give it an id and it hands you that one, leave it out and it's the whole friend group ✨

query string parameters:

  • error_trace
  • from
  • human
  • size

see the get_calendars docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-get-calendar-event.html 💖

🔥 get_calendar_events()

$response = $es->ml->get_calendar_events(
    calendar_id => $id,     # required
)

the get_calendar_events() method retrieves events from a calendar. pull up, we're going through the schedule 📋

query string parameters:

  • end
  • error_trace
  • from
  • human
  • job_id
  • size
  • start

see the get_calendar_events docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-get-calendar-event.html 🌸

🎯 FILTER METHODS 🎯

filters!! the little named lists that keep your ML jobs from spiralling. obsessed 💜

💖 put_filter()

$response = $es->ml->put_filter(
    filter_id   => $id,     # required
    body        => {...}    # required
)

the put_filter() method creates a named filter. named, bestie. it has a whole identity ✨

query string parameters:

  • error_trace
  • human

see the put_filter docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-put-filter.html 🦋

🎀 update_filter()

$response = $es->ml->update_filter(
    filter_id   => $id,     # required
    body        => {...}    # required
)

the update_filter() method updates the description of a filter, adds items, or removes items. glow-up, restock, declutter — pick your era 💅

query string parameters:

  • error_trace
  • human

see the update_filter docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-update-filter.html 🌈

🌸 get_filters()

$response = $es->ml->get_filters(
    filter_id   => $id,     # optional
)

the get_filters() method retrieves a named filter or all filters. one, or the whole lineup 👑

query string parameters:

  • error_trace
  • from
  • human
  • size

see the get_filters docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-get-filters.html 🍓

💀 delete_filter()

$response = $es->ml->delete_filter(
    filter_id   => $id,     # required
)

the delete_filter() method deletes a named filter. poof. that filter? never met her 😭

query string parameters:

  • error_trace
  • human

see the delete_filters docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-delete-filter.html 🧚

🔮 FORECAST METHODS 🔮

manifesting the future, but make it stats ✨

🌟 forecast()

$response = $es->ml->forecast(
    job_id      => $id      # required
)

the forecast() method enables you to create a new forecast. yes bestie, it literally predicts the vibe 📈

see the forecast docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-forecast.html 💜

query string parameters:

  • duration
  • error_trace
  • expires_in
  • human

🦋 delete_forecast()

$response = $es->ml->delete_forecast(
    forecast_id => $id,     # required
    job_id      => $id      # required
)

the delete_forecast() method enables you to delete an existing forecast. un-manifesting, we love to see it 🙈

see the delete_forecast docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-delete-forecast.html 🎀

query string parameters:

  • allow_no_forecasts
  • error_trace
  • human
  • timeout

📸 MODEL SNAPSHOT METHODS 📸

snapshots = the receipts of your ML job's journey. scrapbook core 🌸

🎞️ delete_model_snapshot()

$response = $es->ml->delete_model_snapshot(
    snapshot_id => $id      # required
)

the delete_model_snapshot() method enables you to delete an existing model snapshot. yeah that memory is leaving the chat 💀

see the delete_model_snapshot docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-delete-snapshot.html 🦋

query string parameters:

  • error_trace
  • human

🔍 get_model_snapshots()

$response = $es->ml->get_model_snapshots(
    job_id      => $job_id,         # required
    snapshot_id => $snapshot_id     # optional
)

the get_model_snapshots() method enables you to retrieve information about model snapshots. the job_id is required, the snapshot_id is optional — so you can pull one specific snapshot or the whole camera roll 📷

see the get_model_snapshots docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-get-snapshot.html 💖

query string parameters:

  • desc
  • end
  • error_trace
  • from
  • human
  • size
  • sort
  • start

⏪ revert_model_snapshot()

$response = $es->ml->revert_model_snapshot(
    job_id      => $job_id,         # required
    snapshot_id => $snapshot_id     # required
)

the revert_model_snapshots() method enables you to revert to a specific snapshot. it's giving time travel, no cap ⏳

see the revert_model_snapshot docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-revert-snapshot.html 🌈

query string parameters:

  • delete_intervening_results
  • error_trace
  • human

✏️ update_model_snapshot()

$response = $es->ml->update_model_snapshot(
    job_id      => $job_id,         # required
    snapshot_id => $snapshot_id     # required
)

the update_model_snapshots() method enables you to update certain properties of a snapshot. little tweak, little refresh, still iconic 💅

see the update_model_snapshot docs for more info: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-update-snapshot.html 🍓

query string parameters:

  • error_trace
  • human

💖 RESULT METHODS 💖

okay bestie, this is where the actual tea lives — all the results your jobs cooked up. let's get into it 🔥

get_buckets()

$response = $es->ml->get_buckets(
    job_id      => $job_id,         # required
    timestamp   => $timestamp       # optional
)

the get_buckets() method lets you pull job results for one or more buckets. buckets!! the little time-windows where your anomaly detection does its thing ✨

the full deets live in the get_buckets docs: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-get-bucket.html

query string parameters, no cap:

  • anomaly_score
  • desc
  • end
  • error_trace
  • exclude_interim
  • expand
  • from
  • human
  • size
  • sort
  • start

get_overall_buckets()

$response = $es->ml->get_overall_buckets(
    job_id      => $job_id,         # required
)

the get_overall_buckets() method grabs overall bucket results that summarize the bucket results of multiple jobs. it's giving group-project energy but, like, the good kind 🫶

the get_overall_buckets docs are right here: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-get-overall-buckets.html

query string parameters:

  • allow_no_jobs
  • bucket_span
  • end
  • error_trace
  • exclude_interim
  • human
  • overall_score
  • start
  • top_n

get_categories()

$response = $es->ml->get_categories(
    job_id      => $job_id,         # required
    category_id => $category_id     # optional
)

the get_categories() method lets you retrieve job results for one or more categories. sorting your anomalies into lil categories? iconic, honestly 👑

get_categories docs over here: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-get-category.html

query string parameters:

  • error_trace
  • from
  • human
  • size

get_influencers()

$response = $es->ml->get_influencers(
    job_id      => $job_id,         # required
)

the get_influencers() method lets you retrieve job results for one or more influencers. yes, your data has influencers now. main character fields, periodt 💅

get_influencers docs: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-get-influencer.html

query string parameters:

  • desc
  • end
  • error_trace
  • exclude_interim
  • expand
  • from
  • human
  • influencer_score
  • size
  • sort
  • start

get_records()

$response = $es->ml->get_records(
    job_id      => $job_id,         # required
)

the get_records() method lets you retrieve anomaly records for a job. the receipts, bestie. literally the receipts 🧾✨

get_records docs: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-get-record.html

query string parameters:

  • desc
  • end
  • error_trace
  • exclude_interim
  • expand
  • from
  • human
  • record_score
  • size
  • sort
  • start

🌈 FILE STRUCTURE METHODS 🌈

find_file_structure

$response = $es->ml->find_file_structure(
    body    => { ... },         # required
)

the find_file_structure() method finds the structure of a text file that contains data suitable to be ingested into Elasticsearch. it's like a vibe check, but for your file's schema 🔍

find_file_structure docs: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-find-file-structure.html

query string parameters:

  • charset
  • column_names
  • delimiter
  • error_trace
  • explain
  • format
  • grok_pattern
  • has_header_row
  • human
  • lines_to_sample
  • quote
  • should_trim_fields
  • timeout
  • timestamp_field
  • timestamp_format

⭐ INFO METHODS ⭐

info

$response = $es->ml->info();

the info() method returns the defaults and limits used by machine learning. it's the ML subsystem handing you its whole stat sheet, and we love a transparent king 💜

the docs for this one are (yes, the original calls them this) the find_file_structure docs: https://www.elastic.co/guide/en/elasticsearch/reference/current/get-ml-info.html

query string parameters:

  • error_trace
  • human

🚀 UPGRADE METHODS 🚀

set_upgrade_mode

$response = $es->ml->set_upgrade_mode();

the set_upgrade_mode() method sets a cluster wide upgrade_mode setting that prepares machine learning indices for an upgrade. glow-up time for your indices, fr fr ✨

heads up bestie: this is cluster wide, so it's not a little personal setting — it touches the whole cluster, okay?

set_upgrade_mode docs: https://www.elastic.co/guide/en/elasticsearch/reference/current/ml-set-upgrade-mode.html

query string parameters:

  • enabled
  • error_trace
  • human
  • timeout

👑 AUTHOR 👑

Enrico Zimuel <enrico.zimuel@elastic.co>

the genius behind this whole thing, we stan 🫶

This software is Copyright (c) 2021 by Elasticsearch BV.

This is free software, licensed under:

The Apache License, Version 2.0, January 2004

and that's a wrap on the whole page, bestie!! 🎀 this was documented for perl v5.32.1, dated 2021-10-23, and it's the Search...ct::ML(3pm) page from the User Contributed Perl Documentation. go forth and machine-learn responsibly, ily, bye 💖🦋🌸

📦 where this page comes from