ML Recommendation and streams

ml recommend fit ml recommend fit ml recommend items ml recommend rating ml recommend similar ml recommend evaluate ml sketch create ml sketch create ml sketch add ml sketch query ml sketch result ml sketch cdf ml sketch show ml sketch exist ml sketch delete ml sketch delete_all

ml recommend fit <id> <json_ratings>

Description

Fit a recommender with the default parameters (ALS matrix factorization with biases, 10 factors) from ratings [user, item, rating]

Parameters

id: The model id - String - required
json_ratings: The ratings: [[user, item, rating], ...] - String - required

admin
ml recommend fit "rec" "[[\"ann\", \"A\", 5], [\"ann\", \"B\", 4], [\"ann\", \"C\", 1], [\"bob\", \"A\", 4], [\"bob\", \"B\", 5], [\"bob\", \"D\", 1], [\"cat\", \"C\", 5], [\"cat\", \"D\", 4], [\"cat\", \"A\", 1], [\"dan\", \"C\", 4], [\"dan\", \"D\", 5], [\"dan\", \"B\", 2], [\"eve\", \"A\", 5], [\"eve\", \"C\", 2]]"

mentdb
{"id": "rec", "algo": "recommend_als", "kind": "recommender", "n_samples": 14, "n_features": 0, "params": {}, "method": "als", "users": 5, "items": 4, "global_mean": 3.4285714285714284, "train_rmse": 0.12161316814389235}

ml recommend fit <id> <json_ratings> <json_params>

Description

Fit a recommender with parameters: method als {factors, iterations, lambda, seed} or item_knn {k} (item-item cosine similarity on the ratings centered by the user mean)

Parameters

id: The model id - String - required
json_ratings: The ratings: [[user, item, rating], ...] - String - required
json_params: The parameters - String - required

admin
ml recommend fit "rec" "[[\"ann\", \"A\", 5], [\"ann\", \"B\", 4], [\"ann\", \"C\", 1], [\"bob\", \"A\", 4], [\"bob\", \"B\", 5], [\"bob\", \"D\", 1], [\"cat\", \"C\", 5], [\"cat\", \"D\", 4], [\"cat\", \"A\", 1], [\"dan\", \"C\", 4], [\"dan\", \"D\", 5], [\"dan\", \"B\", 2], [\"eve\", \"A\", 5], [\"eve\", \"C\", 2]]" "{\"method\": \"item_knn\", \"k\": 3}"

mentdb
{"id": "rec", "algo": "recommend_item_knn", "kind": "recommender", "n_samples": 14, "n_features": 0, "params": {"method":"item_knn","k":3}, "method": "item_knn", "users": 5, "items": 4, "global_mean": 3.4285714285714284, "train_rmse": 1.502643173856158}

Used in use cases (1 recipe)

Data scientist › Product recommendations with matrix factorization

Often used with

ml model delete · ml recommend items · ml recommend rating · ml recommend similar

ml recommend items <id> <user> <n>

Description

The n best items not yet rated by a user, with the predicted rating

Parameters

id: The model id - String - required
user: The user - String - required
n: The number of items - Number - required

admin
ml recommend fit "rec" "[[\"ann\", \"A\", 5], [\"ann\", \"B\", 4], [\"ann\", \"C\", 1], [\"bob\", \"A\", 4], [\"bob\", \"B\", 5], [\"bob\", \"D\", 1], [\"cat\", \"C\", 5], [\"cat\", \"D\", 4], [\"cat\", \"A\", 1], [\"dan\", \"C\", 4], [\"dan\", \"D\", 5], [\"dan\", \"B\", 2], [\"eve\", \"A\", 5], [\"eve\", \"C\", 2]]" "{\"factors\": 2}";
ml recommend items "rec" "eve" 2

mentdb
[{"item": "B", "score": 4.208004883696468}, {"item": "D", "score": 2.791156224464721}]

Used in use cases (1 recipe)

Data scientist › Product recommendations with matrix factorization

Often used with

ml model delete · ml recommend fit · ml recommend rating · ml recommend similar

ml recommend rating <id> <user> <item>

Description

Predicted rating of an item by a user

Parameters

id: The model id - String - required
user: The user - String - required
item: The item - String - required

admin
ml recommend fit "rec" "[[\"ann\", \"A\", 5], [\"ann\", \"B\", 4], [\"ann\", \"C\", 1], [\"bob\", \"A\", 4], [\"bob\", \"B\", 5], [\"bob\", \"D\", 1], [\"cat\", \"C\", 5], [\"cat\", \"D\", 4], [\"cat\", \"A\", 1], [\"dan\", \"C\", 4], [\"dan\", \"D\", 5], [\"dan\", \"B\", 2], [\"eve\", \"A\", 5], [\"eve\", \"C\", 2]]" "{\"factors\": 2}";
ml recommend rating "rec" "eve" "B"

mentdb
4.208004883696468

Used in use cases (1 recipe)

Data scientist › Product recommendations with matrix factorization

Often used with

ml model delete · ml recommend fit · ml recommend items · ml recommend similar

ml recommend similar <id> <item> <n>

Description

The n items most similar to an item (cosine of the ALS factors or item-item similarity)

Parameters

id: The model id - String - required
item: The item - String - required
n: The number of items - Number - required

admin
ml recommend fit "rec" "[[\"ann\", \"A\", 5], [\"ann\", \"B\", 4], [\"ann\", \"C\", 1], [\"bob\", \"A\", 4], [\"bob\", \"B\", 5], [\"bob\", \"D\", 1], [\"cat\", \"C\", 5], [\"cat\", \"D\", 4], [\"cat\", \"A\", 1], [\"dan\", \"C\", 4], [\"dan\", \"D\", 5], [\"dan\", \"B\", 2], [\"eve\", \"A\", 5], [\"eve\", \"C\", 2]]" "{\"method\": \"item_knn\"}";
ml recommend similar "rec" "A" 2

mentdb
[{"item": "B", "similarity": 0.27457869476231794}, {"item": "D", "similarity": -0.34006912694340236}]

Used in use cases (1 recipe)

Data scientist › Product recommendations with matrix factorization

Often used with

ml model delete · ml recommend fit · ml recommend items · ml recommend rating

ml recommend evaluate <id> <json_ratings>

Description

RMSE and MAE on test ratings (the unknown users or items are skipped)

Parameters

id: The model id - String - required
json_ratings: The test ratings - String - required

admin
ml recommend fit "rec" "[[\"ann\", \"A\", 5], [\"ann\", \"B\", 4], [\"ann\", \"C\", 1], [\"bob\", \"A\", 4], [\"bob\", \"B\", 5], [\"bob\", \"D\", 1], [\"cat\", \"C\", 5], [\"cat\", \"D\", 4], [\"cat\", \"A\", 1], [\"dan\", \"C\", 4], [\"dan\", \"D\", 5], [\"dan\", \"B\", 2], [\"eve\", \"A\", 5], [\"eve\", \"C\", 2]]" "{\"factors\": 2}";
ml recommend evaluate "rec" "[[\"ann\", \"D\", 2], [\"eve\", \"B\", 4], [\"zoe\", \"A\", 3]]"

mentdb
{"rmse": 0.15767132398454867, "mae": 0.14417227248174513, "n": 2, "skipped": 1}

ml sketch create <id> <type>

Description

Create a sketch with the default parameters.
Types: hll (distinct count, HyperLogLog++), countmin (frequencies), topk (most frequent values, Space-Saving), tdigest (quantiles; exact up to 10 000 values), bloom (membership).
Streams: hoeffding (incremental classifier, Hoeffding tree), adwin (change detection, ADWIN), closed_itemsets (closed itemsets, CloStream), recent_itemsets (recent frequent itemsets with decay, estDec)

Parameters

id: The sketch id - String - required
type: hll, countmin, topk, tdigest, bloom, hoeffding, adwin, closed_itemsets or recent_itemsets - String - required

admin
ml sketch create "visitors" "hll"

mentdb
{"id": "visitors", "type": "hll", "params": {}, "count": 0}

ml sketch create <id> <type> <json_params>

Description

Create a sketch with parameters: hll {precision: 14}, countmin {epsilon: 0.001, confidence: 0.99, seed}, topk {capacity: 1000}, tdigest {compression: 100}, bloom {expected: 100000, fpp: 0.01}, hoeffding {grace_period: 200, delta: 1e-7, tie_threshold: 0.05, max_depth: 20}, adwin {delta: 0.002, max_width: 5000}, closed_itemsets {min_support: 2}, recent_itemsets {min_support: 0.1, min_significance: 0.05, decay_base: 2, half_life: 1000}

Parameters

id: The sketch id - String - required
type: The type - String - required
json_params: The parameters - String - required

admin
ml sketch create "seen" "bloom" "{\"expected\": 1000000, \"fpp\": 0.001}"

mentdb
{"id": "seen", "type": "bloom", "params": {"expected":1000000,"fpp":0.001}, "count": 0}

Used in use cases (2 recipes)

IoT & sensor engineer › Quantiles and distinct devices of a telemetry stream
IoT & sensor engineer › Detect a change in a sensor stream

Often used with

ml sketch add · ml sketch delete · ml sketch query · ml sketch result · ml sketch cdf

ml sketch add <id> <values>

Description

Add a JSON array of values (or one value) to a sketch; returns the number of values added since the creation.
Streams: hoeffding = rows [x1, x2, ..., label], adwin = numbers, closed_itemsets and recent_itemsets = transactions (arrays of items)

Parameters

id: The sketch id - String - required
values: A JSON array of values or one value - String - required

admin
ml sketch create "visitors" "hll";
ml sketch add "visitors" "[\"10.0.0.1\", \"10.0.0.2\", \"10.0.0.1\"]"

mentdb
3

Used in use cases (2 recipes)

IoT & sensor engineer › Quantiles and distinct devices of a telemetry stream
IoT & sensor engineer › Detect a change in a sensor stream

Often used with

ml sketch create · ml sketch delete · ml sketch query · ml sketch result · ml sketch cdf

ml sketch query <id> <value>

Description

Query a sketch: countmin = estimated frequency of a value, bloom = 1 if the value may be present / 0 if it is surely absent, tdigest = quantile q in [0, 1], topk = the k most frequent values; hoeffding = the label of a row (or of rows), "rules" = the rules of the tree; adwin = "drift" (1 if the last value detected a change) or "mean"; closed_itemsets = the closed itemsets with a minimum support

Parameters

id: The sketch id - String - required
value: The value, q or k - String - required

admin
ml sketch create "words" "countmin";
ml sketch add "words" "[\"pump\", \"valve\", \"pump\"]";
ml sketch query "words" "pump"

mentdb
2

Most frequent

admin
ml sketch create "top" "topk";
ml sketch add "top" "[\"E12\", \"E07\", \"E12\", \"E03\", \"E12\", \"E07\"]";
ml sketch query "top" 2

mentdb
[{"value": "E12", "count": 3, "error": 0}, {"value": "E07", "count": 2, "error": 0}]

Quantile 95%

admin
ml sketch create "lat" "tdigest";
ml sketch add "lat" "[12, 15, 11, 19, 14, 120, 13, 16, 12, 18]";
ml sketch query "lat" 0.95

mentdb
74.5499999999999

Hoeffding tree: learn a stream of rows, then classify

admin
ml sketch create "ht" "hoeffding" "{\"grace_period\": 50}";
ml sketch add "ht" "[[6.23, 7.42, \"alarm\"], [7.95, 9.42, \"alarm\"], [7.4, 9.22, \"alarm\"], [0.29, 4.66, \"ok\"], [9.43, 6.49, \"alarm\"], [9.01, 1.13, \"alarm\"], [4.69, 2.47, \"ok\"], [5.44, 5.74, \"alarm\"], [0.13, 2.17, \"ok\"], [2.79, 9.16, \"alarm\"], [7.66, 1.6, \"ok\"], [7.97, 1.39, \"ok\"], [6.17, 1.27, \"ok\"], [0.02, 8.71, \"ok\"], [2.09, 2.15, \"ok\"], [9.82, 8.72, \"alarm\"], [2.89, 9.61, \"alarm\"], [5.39, 6.78, \"alarm\"], [2.05, 9.41, \"alarm\"], [6.91, 9.67, \"alarm\"], [8.94, 2.99, \"alarm\"], [3.61, 1.66, \"ok\"], [1.46, 0.65, \"ok\"], [3.01, 6.03, \"ok\"], [0.03, 6.78, \"ok\"], [3.38, 3.1, \"ok\"], [8.19, 4.81, \"alarm\"], [3.16, 4.81, \"ok\"], [7.05, 0.57, \"ok\"], [9.75, 0.23, \"ok\"], [7.5, 8.45, \"alarm\"], [0.18, 7.88, \"ok\"], [3.66, 5.79, \"ok\"], [0.09, 0.47, \"ok\"], [1.81, 9.55, \"alarm\"], [1.97, 7.56, \"ok\"], [9.3, 9.42, \"alarm\"], [3.44, 3.55, \"ok\"], [5.25, 7.76, \"alarm\"], [1.08, 7.48, \"ok\"], [7.97, 8.6, \"alarm\"], [0.37, 9.46, \"ok\"], [0.91, 3.41, \"ok\"], [6.11, 9.18, \"alarm\"], [3.4, 9.24, \"alarm\"], [5.45, 3.12, \"ok\"], [3.17, 1.77, \"ok\"], [0.78, 1.49, \"ok\"], [6.89, 9.97, \"alarm\"], [1.62, 0.49, \"ok\"], [9.87, 5.34, \"alarm\"], [4.06, 2.37, \"ok\"], [5.94, 8.26, \"alarm\"], [4.56, 4.22, \"ok\"], [0.56, 9.16, \"ok\"], [0.33, 4.94, \"ok\"], [8.38, 1.31, \"ok\"], [7.32, 9.5, \"alarm\"], [6.3, 7.88, \"alarm\"], [1.07, 4.35, \"ok\"], [1.49, 8.45, \"ok\"], [2.95, 4.53, \"ok\"], [9.99, 8.52, \"alarm\"], [9.76, 4.54, \"alarm\"], [4.88, 7.3, \"alarm\"], [4.79, 2.91, \"ok\"], [4.04, 1.47, \"ok\"], [3.77, 9.88, \"alarm\"], [9.6, 6.27, \"alarm\"], [4.99, 3.38, \"ok\"], [0.89, 2.72, \"ok\"], [7.82, 8.67, \"alarm\"], [3.61, 7.86, \"alarm\"], [7.75, 6.95, \"alarm\"], [6.64, 7.6, \"alarm\"], [3.63, 7.04, \"alarm\"], [2.81, 4.86, \"ok\"], [7.7, 6.91, \"alarm\"], [2.94, 9.46, \"alarm\"], [6.5, 5.81, \"alarm\"], [0.12, 5.47, \"ok\"], [2.51, 6.72, \"ok\"], [4.63, 8.17, \"alarm\"], [6.47, 7.98, \"alarm\"], [3.48, 6.44, \"ok\"], [7.38, 8.28, \"alarm\"], [3.5, 8.43, \"alarm\"], [8.7, 6.88, \"alarm\"], [9.76, 9.57, \"alarm\"], [5.18, 5.29, \"alarm\"], [1.66, 8.37, \"alarm\"], [9.37, 4.77, \"alarm\"], [6.91, 7.2, \"alarm\"], [7.3, 1.72, \"ok\"], [7.8, 5.81, \"alarm\"], [6.66, 4.21, \"alarm\"], [6.24, 7.75, \"alarm\"], [6.37, 7.2, \"alarm\"], [0.28, 1.6, \"ok\"], [4.41, 6.5, \"alarm\"], [2.19, 6.86, \"ok\"], [6.31, 0.42, \"ok\"], [4.72, 2.26, \"ok\"], [0.54, 1.34, \"ok\"], [3.17, 1.82, \"ok\"], [1.93, 0.36, \"ok\"], [4.65, 3.8, \"ok\"], [6.12, 5.9, \"alarm\"], [2.38, 9.03, \"alarm\"], [0.01, 4.05, \"ok\"], [2.79, 4.1, \"ok\"], [1.15, 8.31, \"ok\"], [3.74, 0.36, \"ok\"], [6.14, 0.95, \"ok\"], [5.45, 3.39, \"ok\"], [5.81, 9.58, \"alarm\"], [8.19, 4.19, \"alarm\"], [8.13, 6.42, \"alarm\"], [3.69, 1.42, \"ok\"], [5.96, 5.64, \"alarm\"], [9.57, 9.68, \"alarm\"], [6.09, 3.51, \"ok\"], [8.93, 0.01, \"ok\"], [1.08, 5.66, \"ok\"], [6.15, 1.41, \"ok\"], [6.29, 8.91, \"alarm\"], [3.76, 4.32, \"ok\"], [2.26, 2.91, \"ok\"], [9.72, 3.8, \"alarm\"], [9.61, 9.14, \"alarm\"], [5.96, 2.6, \"ok\"], [9.81, 4.96, \"alarm\"], [4.15, 3.19, \"ok\"], [9.84, 4.92, \"alarm\"], [2.86, 4.77, \"ok\"], [1.22, 6.22, \"ok\"], [4.43, 2.93, \"ok\"], [7.82, 8.27, \"alarm\"], [0.13, 5.33, \"ok\"], [2.74, 9.35, \"alarm\"], [7.82, 2.46, \"alarm\"], [2.68, 1.55, \"ok\"], [9.89, 2.93, \"alarm\"], [6.08, 4.75, \"alarm\"], [6.45, 6.04, \"alarm\"], [7.43, 1.18, \"ok\"], [7.6, 3.01, \"alarm\"], [5.33, 3.36, \"ok\"], [2.97, 5.3, \"ok\"], [4.64, 3.61, \"ok\"], [7.45, 5.91, \"alarm\"], [0.36, 2.52, \"ok\"], [4.56, 9.17, \"alarm\"], [8.88, 5.46, \"alarm\"], [0.15, 7.78, \"ok\"], [4.28, 5.76, \"alarm\"], [7.08, 6.32, \"alarm\"], [4.82, 9.12, \"alarm\"], [3.85, 3.92, \"ok\"], [8.52, 1.96, \"alarm\"], [2.96, 8.3, \"alarm\"], [0.66, 8.36, \"ok\"], [6.95, 4.33, \"alarm\"], [2.86, 7.81, \"alarm\"], [9.11, 1.43, \"alarm\"], [4.78, 5.49, \"alarm\"], [4.98, 3.31, \"ok\"], [1.54, 5.86, \"ok\"], [8.12, 0.68, \"ok\"], [2.3, 8.2, \"alarm\"], [7.92, 6.64, \"alarm\"], [0.26, 7.23, \"ok\"], [9.79, 9.98, \"alarm\"], [7.01, 0.49, \"ok\"], [8.42, 2.19, \"alarm\"], [6.46, 9.52, \"alarm\"], [7.12, 1.35, \"ok\"], [2.92, 9.18, \"alarm\"], [1.5, 6.11, \"ok\"], [4.14, 1.61, \"ok\"], [6.22, 0.44, \"ok\"], [1.08, 3.79, \"ok\"], [0.72, 0.58, \"ok\"], [5.75, 7.42, \"alarm\"], [8.78, 1.34, \"alarm\"], [4.32, 3.15, \"ok\"], [6.0, 4.9, \"alarm\"], [9.39, 3.74, \"alarm\"], [0.56, 6.97, \"ok\"], [1.51, 6.31, \"ok\"], [5.06, 9.1, \"alarm\"], [5.55, 6.21, \"alarm\"], [2.63, 5.52, \"ok\"], [2.54, 7.51, \"alarm\"], [5.17, 1.34, \"ok\"], [2.34, 3.71, \"ok\"], [7.37, 1.79, \"ok\"], [7.13, 6.55, \"alarm\"], [0.85, 6.68, \"ok\"], [0.91, 1.25, \"ok\"], [5.94, 2.39, \"ok\"], [8.77, 4.8, \"alarm\"], [3.23, 7.96, \"alarm\"], [0.29, 7.25, \"ok\"], [0.54, 1.51, \"ok\"], [9.52, 6.81, \"alarm\"], [2.23, 1.16, \"ok\"], [9.73, 6.65, \"alarm\"], [8.21, 1.4, \"ok\"], [6.25, 3.54, \"ok\"], [2.35, 3.33, \"ok\"], [6.14, 3.49, \"ok\"], [3.86, 1.36, \"ok\"], [8.31, 6.48, \"alarm\"], [8.05, 4.33, \"alarm\"], [8.52, 5.17, \"alarm\"], [5.93, 5.73, \"alarm\"], [7.4, 3.96, \"alarm\"], [0.97, 0.33, \"ok\"], [2.02, 0.39, \"ok\"], [8.89, 4.81, \"alarm\"], [7.6, 0.0, \"ok\"], [4.7, 8.9, \"alarm\"], [6.19, 4.29, \"alarm\"], [4.66, 1.0, \"ok\"], [1.55, 1.59, \"ok\"], [3.75, 3.86, \"ok\"], [8.8, 1.52, \"alarm\"], [2.54, 2.77, \"ok\"], [1.62, 2.87, \"ok\"], [2.35, 4.82, \"ok\"], [0.32, 9.25, \"ok\"], [3.69, 9.38, \"alarm\"], [6.88, 6.74, \"alarm\"], [4.72, 9.45, \"alarm\"], [1.18, 6.69, \"ok\"], [2.91, 6.74, \"ok\"], [7.29, 1.63, \"ok\"], [2.01, 0.25, \"ok\"], [2.3, 0.78, \"ok\"], [4.01, 9.74, \"alarm\"], [3.64, 3.12, \"ok\"], [4.68, 2.83, \"ok\"], [7.32, 7.18, \"alarm\"], [1.63, 2.41, \"ok\"], [6.72, 9.4, \"alarm\"], [6.46, 4.31, \"alarm\"], [9.75, 0.06, \"ok\"], [0.61, 7.79, \"ok\"], [4.1, 0.45, \"ok\"], [5.48, 9.9, \"alarm\"], [5.19, 3.5, \"ok\"], [0.94, 0.71, \"ok\"], [8.99, 4.91, \"alarm\"], [9.36, 0.54, \"ok\"], [2.43, 0.5, \"ok\"], [3.97, 0.6, \"ok\"], [2.55, 4.07, \"ok\"], [3.06, 0.51, \"ok\"], [0.38, 9.72, \"alarm\"], [1.79, 5.09, \"ok\"], [4.02, 5.31, \"ok\"], [0.84, 3.14, \"ok\"], [1.08, 5.42, \"ok\"], [9.21, 5.98, \"alarm\"], [8.57, 2.14, \"alarm\"], [0.17, 5.4, \"ok\"], [4.87, 5.71, \"alarm\"], [3.77, 6.25, \"alarm\"], [7.26, 9.13, \"alarm\"], [3.08, 4.49, \"ok\"], [8.26, 2.24, \"alarm\"], [1.16, 3.12, \"ok\"], [0.88, 7.72, \"ok\"], [8.19, 3.13, \"alarm\"], [1.3, 0.82, \"ok\"], [2.45, 0.84, \"ok\"], [4.29, 5.77, \"alarm\"], [2.45, 0.61, \"ok\"], [7.01, 0.48, \"ok\"], [2.0, 2.86, \"ok\"], [3.74, 0.99, \"ok\"], [4.21, 3.14, \"ok\"], [7.52, 5.56, \"alarm\"], [8.96, 6.54, \"alarm\"], [7.6, 5.75, \"alarm\"], [4.42, 8.17, \"alarm\"], [6.55, 9.55, \"alarm\"], [7.28, 7.01, \"alarm\"], [2.68, 8.12, \"alarm\"], [3.82, 1.3, \"ok\"], [0.66, 1.69, \"ok\"], [2.63, 6.76, \"ok\"], [2.86, 0.64, \"ok\"], [7.64, 5.57, \"alarm\"], [0.27, 0.51, \"ok\"], [1.3, 3.57, \"ok\"], [8.6, 9.49, \"alarm\"], [6.12, 2.31, \"ok\"], [4.29, 3.62, \"ok\"], [3.3, 0.12, \"ok\"], [5.84, 8.27, \"alarm\"], [7.27, 0.95, \"ok\"], [5.3, 1.71, \"ok\"], [7.09, 4.46, \"alarm\"], [9.39, 7.95, \"alarm\"], [1.18, 3.17, \"ok\"], [9.15, 4.61, \"alarm\"], [4.34, 4.41, \"ok\"], [7.68, 9.36, \"alarm\"], [5.33, 9.7, \"alarm\"], [5.96, 1.04, \"ok\"], [8.14, 4.19, \"alarm\"], [0.52, 9.78, \"alarm\"], [0.33, 5.77, \"ok\"], [4.82, 8.82, \"alarm\"], [3.93, 2.16, \"ok\"], [2.77, 2.01, \"ok\"], [5.62, 3.57, \"ok\"], [7.51, 2.42, \"ok\"], [3.52, 2.48, \"ok\"], [9.82, 8.4, \"alarm\"], [8.5, 6.18, \"alarm\"], [4.01, 1.43, \"ok\"], [8.32, 4.9, \"alarm\"], [0.38, 1.7, \"ok\"], [0.99, 7.18, \"ok\"], [9.0, 1.99, \"alarm\"], [7.97, 3.24, \"alarm\"], [6.83, 8.63, \"alarm\"], [6.23, 8.0, \"alarm\"], [3.76, 0.1, \"ok\"], [5.12, 5.87, \"alarm\"], [1.85, 3.89, \"ok\"], [3.17, 0.27, \"ok\"], [3.12, 3.83, \"ok\"], [4.76, 7.03, \"alarm\"], [3.99, 9.82, \"alarm\"], [8.16, 9.24, \"alarm\"], [6.93, 6.7, \"alarm\"], [5.37, 7.99, \"alarm\"], [3.63, 5.94, \"ok\"], [6.79, 5.22, \"alarm\"], [2.84, 0.78, \"ok\"], [0.87, 3.56, \"ok\"], [5.8, 7.6, \"alarm\"], [7.14, 3.07, \"alarm\"], [9.29, 2.75, \"alarm\"], [7.16, 0.72, \"ok\"], [7.53, 6.7, \"alarm\"], [9.57, 8.97, \"alarm\"], [6.88, 8.38, \"alarm\"], [7.39, 6.11, \"alarm\"], [2.09, 5.17, \"ok\"], [8.96, 2.39, \"alarm\"], [9.75, 5.44, \"alarm\"], [3.93, 0.03, \"ok\"], [3.9, 1.78, \"ok\"], [6.53, 9.0, \"alarm\"], [9.1, 6.14, \"alarm\"], [3.87, 1.09, \"ok\"], [6.89, 5.53, \"alarm\"], [7.15, 3.74, \"alarm\"], [8.87, 2.13, \"alarm\"], [3.06, 2.68, \"ok\"], [6.22, 8.97, \"alarm\"], [1.67, 6.33, \"ok\"], [7.66, 2.25, \"ok\"], [0.63, 5.66, \"ok\"], [8.3, 8.79, \"alarm\"], [7.28, 4.17, \"alarm\"], [4.25, 5.29, \"ok\"], [9.05, 3.02, \"alarm\"], [2.81, 6.05, \"ok\"], [9.67, 1.87, \"alarm\"], [0.3, 1.16, \"ok\"], [5.63, 6.03, \"alarm\"], [1.84, 1.9, \"ok\"], [5.95, 6.46, \"alarm\"], [6.9, 7.29, \"alarm\"], [0.61, 4.85, \"ok\"], [8.35, 9.57, \"alarm\"], [3.18, 8.5, \"alarm\"], [6.45, 9.27, \"alarm\"], [2.29, 6.97, \"ok\"], [8.41, 4.77, \"alarm\"], [1.01, 1.94, \"ok\"], [1.57, 0.93, \"ok\"], [1.52, 6.16, \"ok\"], [1.05, 7.58, \"ok\"], [7.36, 8.16, \"alarm\"], [8.04, 7.13, \"alarm\"], [9.24, 9.57, \"alarm\"], [6.25, 9.65, \"alarm\"], [1.03, 1.03, \"ok\"], [0.64, 2.06, \"ok\"], [3.74, 4.59, \"ok\"], [6.81, 7.43, \"alarm\"], [0.84, 4.03, \"ok\"], [5.43, 3.75, \"ok\"]]";
ml sketch query "ht" "[[2, 3], [8, 7]]"

mentdb
["ok", "alarm"]

ADWIN: the mean jumps from 10 to 14 at the 61st value

admin
ml sketch create "dr" "adwin";
ml sketch add "dr" "[11.0, 10.91, 8.93, 11.08, 7.44, 10.07, 10.36, 9.87, 10.1, 8.4, 9.86, 10.48, 9.07, 10.12, 9.82, 7.77, 9.39, 10.56, 8.46, 10.01, 10.32, 8.73, 9.74, 10.87, 10.7, 9.01, 10.99, 10.33, 10.11, 11.48, 10.71, 10.35, 10.94, 10.65, 11.24, 9.88, 8.24, 11.53, 10.79, 10.51, 10.04, 9.91, 10.09, 9.06, 10.76, 10.08, 9.05, 10.25, 9.67, 9.97, 10.53, 12.65, 8.94, 10.4, 10.7, 10.41, 11.46, 9.57, 9.67, 9.98, 12.64, 15.08, 12.77, 14.54, 15.18, 13.17, 13.82, 13.8, 15.69, 14.26, 12.26, 15.06, 15.02, 12.02, 11.3, 16.58, 14.08, 11.89, 15.04, 15.4, 11.64, 15.26, 14.7, 13.97, 13.3, 14.38, 13.19, 14.15, 13.74, 13.56, 14.54, 14.97, 14.88, 14.44, 15.66, 12.75, 14.06, 14.18, 15.2, 11.76]";
ml sketch query "dr" "mean"

mentdb
13.998249999999999

Used in use cases (2 recipes)

IoT & sensor engineer › Quantiles and distinct devices of a telemetry stream
IoT & sensor engineer › Detect a change in a sensor stream

Often used with

ml sketch add · ml sketch create · ml sketch delete · ml sketch result · ml sketch cdf

ml sketch result <id>

Description

Summary of a sketch: count, distinct values (hll), quantiles (tdigest), top 10 (topk), error (countmin), buckets (bloom), nodes, depth and prequential accuracy (hoeffding), window, mean and changes (adwin), itemsets (closed_itemsets, recent_itemsets)

Parameters

id: The sketch id - String - required

admin
ml sketch create "lat" "tdigest";
ml sketch add "lat" "[12, 15, 11, 19, 14, 120, 13, 16, 12, 18]";
ml sketch result "lat"

mentdb
{"count": 10, "mode": "exact", "min": 11, "p01": 11.09, "p05": 11.45, "p25": 12.25, "median": 14.5, "p75": 17.5, "p95": 74.5499999999999, "p99": 110.91000000000001, "max": 120}

Hoeffding tree: size and prequential accuracy (test then train)

admin
ml sketch create "ht" "hoeffding" "{\"grace_period\": 50}";
ml sketch add "ht" "[[6.23, 7.42, \"alarm\"], [7.95, 9.42, \"alarm\"], [7.4, 9.22, \"alarm\"], [0.29, 4.66, \"ok\"], [9.43, 6.49, \"alarm\"], [9.01, 1.13, \"alarm\"], [4.69, 2.47, \"ok\"], [5.44, 5.74, \"alarm\"], [0.13, 2.17, \"ok\"], [2.79, 9.16, \"alarm\"], [7.66, 1.6, \"ok\"], [7.97, 1.39, \"ok\"], [6.17, 1.27, \"ok\"], [0.02, 8.71, \"ok\"], [2.09, 2.15, \"ok\"], [9.82, 8.72, \"alarm\"], [2.89, 9.61, \"alarm\"], [5.39, 6.78, \"alarm\"], [2.05, 9.41, \"alarm\"], [6.91, 9.67, \"alarm\"], [8.94, 2.99, \"alarm\"], [3.61, 1.66, \"ok\"], [1.46, 0.65, \"ok\"], [3.01, 6.03, \"ok\"], [0.03, 6.78, \"ok\"], [3.38, 3.1, \"ok\"], [8.19, 4.81, \"alarm\"], [3.16, 4.81, \"ok\"], [7.05, 0.57, \"ok\"], [9.75, 0.23, \"ok\"], [7.5, 8.45, \"alarm\"], [0.18, 7.88, \"ok\"], [3.66, 5.79, \"ok\"], [0.09, 0.47, \"ok\"], [1.81, 9.55, \"alarm\"], [1.97, 7.56, \"ok\"], [9.3, 9.42, \"alarm\"], [3.44, 3.55, \"ok\"], [5.25, 7.76, \"alarm\"], [1.08, 7.48, \"ok\"], [7.97, 8.6, \"alarm\"], [0.37, 9.46, \"ok\"], [0.91, 3.41, \"ok\"], [6.11, 9.18, \"alarm\"], [3.4, 9.24, \"alarm\"], [5.45, 3.12, \"ok\"], [3.17, 1.77, \"ok\"], [0.78, 1.49, \"ok\"], [6.89, 9.97, \"alarm\"], [1.62, 0.49, \"ok\"], [9.87, 5.34, \"alarm\"], [4.06, 2.37, \"ok\"], [5.94, 8.26, \"alarm\"], [4.56, 4.22, \"ok\"], [0.56, 9.16, \"ok\"], [0.33, 4.94, \"ok\"], [8.38, 1.31, \"ok\"], [7.32, 9.5, \"alarm\"], [6.3, 7.88, \"alarm\"], [1.07, 4.35, \"ok\"], [1.49, 8.45, \"ok\"], [2.95, 4.53, \"ok\"], [9.99, 8.52, \"alarm\"], [9.76, 4.54, \"alarm\"], [4.88, 7.3, \"alarm\"], [4.79, 2.91, \"ok\"], [4.04, 1.47, \"ok\"], [3.77, 9.88, \"alarm\"], [9.6, 6.27, \"alarm\"], [4.99, 3.38, \"ok\"], [0.89, 2.72, \"ok\"], [7.82, 8.67, \"alarm\"], [3.61, 7.86, \"alarm\"], [7.75, 6.95, \"alarm\"], [6.64, 7.6, \"alarm\"], [3.63, 7.04, \"alarm\"], [2.81, 4.86, \"ok\"], [7.7, 6.91, \"alarm\"], [2.94, 9.46, \"alarm\"], [6.5, 5.81, \"alarm\"], [0.12, 5.47, \"ok\"], [2.51, 6.72, \"ok\"], [4.63, 8.17, \"alarm\"], [6.47, 7.98, \"alarm\"], [3.48, 6.44, \"ok\"], [7.38, 8.28, \"alarm\"], [3.5, 8.43, \"alarm\"], [8.7, 6.88, \"alarm\"], [9.76, 9.57, \"alarm\"], [5.18, 5.29, \"alarm\"], [1.66, 8.37, \"alarm\"], [9.37, 4.77, \"alarm\"], [6.91, 7.2, \"alarm\"], [7.3, 1.72, \"ok\"], [7.8, 5.81, \"alarm\"], [6.66, 4.21, \"alarm\"], [6.24, 7.75, \"alarm\"], [6.37, 7.2, \"alarm\"], [0.28, 1.6, \"ok\"], [4.41, 6.5, \"alarm\"], [2.19, 6.86, \"ok\"], [6.31, 0.42, \"ok\"], [4.72, 2.26, \"ok\"], [0.54, 1.34, \"ok\"], [3.17, 1.82, \"ok\"], [1.93, 0.36, \"ok\"], [4.65, 3.8, \"ok\"], [6.12, 5.9, \"alarm\"], [2.38, 9.03, \"alarm\"], [0.01, 4.05, \"ok\"], [2.79, 4.1, \"ok\"], [1.15, 8.31, \"ok\"], [3.74, 0.36, \"ok\"], [6.14, 0.95, \"ok\"], [5.45, 3.39, \"ok\"], [5.81, 9.58, \"alarm\"], [8.19, 4.19, \"alarm\"], [8.13, 6.42, \"alarm\"], [3.69, 1.42, \"ok\"], [5.96, 5.64, \"alarm\"], [9.57, 9.68, \"alarm\"], [6.09, 3.51, \"ok\"], [8.93, 0.01, \"ok\"], [1.08, 5.66, \"ok\"], [6.15, 1.41, \"ok\"], [6.29, 8.91, \"alarm\"], [3.76, 4.32, \"ok\"], [2.26, 2.91, \"ok\"], [9.72, 3.8, \"alarm\"], [9.61, 9.14, \"alarm\"], [5.96, 2.6, \"ok\"], [9.81, 4.96, \"alarm\"], [4.15, 3.19, \"ok\"], [9.84, 4.92, \"alarm\"], [2.86, 4.77, \"ok\"], [1.22, 6.22, \"ok\"], [4.43, 2.93, \"ok\"], [7.82, 8.27, \"alarm\"], [0.13, 5.33, \"ok\"], [2.74, 9.35, \"alarm\"], [7.82, 2.46, \"alarm\"], [2.68, 1.55, \"ok\"], [9.89, 2.93, \"alarm\"], [6.08, 4.75, \"alarm\"], [6.45, 6.04, \"alarm\"], [7.43, 1.18, \"ok\"], [7.6, 3.01, \"alarm\"], [5.33, 3.36, \"ok\"], [2.97, 5.3, \"ok\"], [4.64, 3.61, \"ok\"], [7.45, 5.91, \"alarm\"], [0.36, 2.52, \"ok\"], [4.56, 9.17, \"alarm\"], [8.88, 5.46, \"alarm\"], [0.15, 7.78, \"ok\"], [4.28, 5.76, \"alarm\"], [7.08, 6.32, \"alarm\"], [4.82, 9.12, \"alarm\"], [3.85, 3.92, \"ok\"], [8.52, 1.96, \"alarm\"], [2.96, 8.3, \"alarm\"], [0.66, 8.36, \"ok\"], [6.95, 4.33, \"alarm\"], [2.86, 7.81, \"alarm\"], [9.11, 1.43, \"alarm\"], [4.78, 5.49, \"alarm\"], [4.98, 3.31, \"ok\"], [1.54, 5.86, \"ok\"], [8.12, 0.68, \"ok\"], [2.3, 8.2, \"alarm\"], [7.92, 6.64, \"alarm\"], [0.26, 7.23, \"ok\"], [9.79, 9.98, \"alarm\"], [7.01, 0.49, \"ok\"], [8.42, 2.19, \"alarm\"], [6.46, 9.52, \"alarm\"], [7.12, 1.35, \"ok\"], [2.92, 9.18, \"alarm\"], [1.5, 6.11, \"ok\"], [4.14, 1.61, \"ok\"], [6.22, 0.44, \"ok\"], [1.08, 3.79, \"ok\"], [0.72, 0.58, \"ok\"], [5.75, 7.42, \"alarm\"], [8.78, 1.34, \"alarm\"], [4.32, 3.15, \"ok\"], [6.0, 4.9, \"alarm\"], [9.39, 3.74, \"alarm\"], [0.56, 6.97, \"ok\"], [1.51, 6.31, \"ok\"], [5.06, 9.1, \"alarm\"], [5.55, 6.21, \"alarm\"], [2.63, 5.52, \"ok\"], [2.54, 7.51, \"alarm\"], [5.17, 1.34, \"ok\"], [2.34, 3.71, \"ok\"], [7.37, 1.79, \"ok\"], [7.13, 6.55, \"alarm\"], [0.85, 6.68, \"ok\"], [0.91, 1.25, \"ok\"], [5.94, 2.39, \"ok\"], [8.77, 4.8, \"alarm\"], [3.23, 7.96, \"alarm\"], [0.29, 7.25, \"ok\"], [0.54, 1.51, \"ok\"], [9.52, 6.81, \"alarm\"], [2.23, 1.16, \"ok\"], [9.73, 6.65, \"alarm\"], [8.21, 1.4, \"ok\"], [6.25, 3.54, \"ok\"], [2.35, 3.33, \"ok\"], [6.14, 3.49, \"ok\"], [3.86, 1.36, \"ok\"], [8.31, 6.48, \"alarm\"], [8.05, 4.33, \"alarm\"], [8.52, 5.17, \"alarm\"], [5.93, 5.73, \"alarm\"], [7.4, 3.96, \"alarm\"], [0.97, 0.33, \"ok\"], [2.02, 0.39, \"ok\"], [8.89, 4.81, \"alarm\"], [7.6, 0.0, \"ok\"], [4.7, 8.9, \"alarm\"], [6.19, 4.29, \"alarm\"], [4.66, 1.0, \"ok\"], [1.55, 1.59, \"ok\"], [3.75, 3.86, \"ok\"], [8.8, 1.52, \"alarm\"], [2.54, 2.77, \"ok\"], [1.62, 2.87, \"ok\"], [2.35, 4.82, \"ok\"], [0.32, 9.25, \"ok\"], [3.69, 9.38, \"alarm\"], [6.88, 6.74, \"alarm\"], [4.72, 9.45, \"alarm\"], [1.18, 6.69, \"ok\"], [2.91, 6.74, \"ok\"], [7.29, 1.63, \"ok\"], [2.01, 0.25, \"ok\"], [2.3, 0.78, \"ok\"], [4.01, 9.74, \"alarm\"], [3.64, 3.12, \"ok\"], [4.68, 2.83, \"ok\"], [7.32, 7.18, \"alarm\"], [1.63, 2.41, \"ok\"], [6.72, 9.4, \"alarm\"], [6.46, 4.31, \"alarm\"], [9.75, 0.06, \"ok\"], [0.61, 7.79, \"ok\"], [4.1, 0.45, \"ok\"], [5.48, 9.9, \"alarm\"], [5.19, 3.5, \"ok\"], [0.94, 0.71, \"ok\"], [8.99, 4.91, \"alarm\"], [9.36, 0.54, \"ok\"], [2.43, 0.5, \"ok\"], [3.97, 0.6, \"ok\"], [2.55, 4.07, \"ok\"], [3.06, 0.51, \"ok\"], [0.38, 9.72, \"alarm\"], [1.79, 5.09, \"ok\"], [4.02, 5.31, \"ok\"], [0.84, 3.14, \"ok\"], [1.08, 5.42, \"ok\"], [9.21, 5.98, \"alarm\"], [8.57, 2.14, \"alarm\"], [0.17, 5.4, \"ok\"], [4.87, 5.71, \"alarm\"], [3.77, 6.25, \"alarm\"], [7.26, 9.13, \"alarm\"], [3.08, 4.49, \"ok\"], [8.26, 2.24, \"alarm\"], [1.16, 3.12, \"ok\"], [0.88, 7.72, \"ok\"], [8.19, 3.13, \"alarm\"], [1.3, 0.82, \"ok\"], [2.45, 0.84, \"ok\"], [4.29, 5.77, \"alarm\"], [2.45, 0.61, \"ok\"], [7.01, 0.48, \"ok\"], [2.0, 2.86, \"ok\"], [3.74, 0.99, \"ok\"], [4.21, 3.14, \"ok\"], [7.52, 5.56, \"alarm\"], [8.96, 6.54, \"alarm\"], [7.6, 5.75, \"alarm\"], [4.42, 8.17, \"alarm\"], [6.55, 9.55, \"alarm\"], [7.28, 7.01, \"alarm\"], [2.68, 8.12, \"alarm\"], [3.82, 1.3, \"ok\"], [0.66, 1.69, \"ok\"], [2.63, 6.76, \"ok\"], [2.86, 0.64, \"ok\"], [7.64, 5.57, \"alarm\"], [0.27, 0.51, \"ok\"], [1.3, 3.57, \"ok\"], [8.6, 9.49, \"alarm\"], [6.12, 2.31, \"ok\"], [4.29, 3.62, \"ok\"], [3.3, 0.12, \"ok\"], [5.84, 8.27, \"alarm\"], [7.27, 0.95, \"ok\"], [5.3, 1.71, \"ok\"], [7.09, 4.46, \"alarm\"], [9.39, 7.95, \"alarm\"], [1.18, 3.17, \"ok\"], [9.15, 4.61, \"alarm\"], [4.34, 4.41, \"ok\"], [7.68, 9.36, \"alarm\"], [5.33, 9.7, \"alarm\"], [5.96, 1.04, \"ok\"], [8.14, 4.19, \"alarm\"], [0.52, 9.78, \"alarm\"], [0.33, 5.77, \"ok\"], [4.82, 8.82, \"alarm\"], [3.93, 2.16, \"ok\"], [2.77, 2.01, \"ok\"], [5.62, 3.57, \"ok\"], [7.51, 2.42, \"ok\"], [3.52, 2.48, \"ok\"], [9.82, 8.4, \"alarm\"], [8.5, 6.18, \"alarm\"], [4.01, 1.43, \"ok\"], [8.32, 4.9, \"alarm\"], [0.38, 1.7, \"ok\"], [0.99, 7.18, \"ok\"], [9.0, 1.99, \"alarm\"], [7.97, 3.24, \"alarm\"], [6.83, 8.63, \"alarm\"], [6.23, 8.0, \"alarm\"], [3.76, 0.1, \"ok\"], [5.12, 5.87, \"alarm\"], [1.85, 3.89, \"ok\"], [3.17, 0.27, \"ok\"], [3.12, 3.83, \"ok\"], [4.76, 7.03, \"alarm\"], [3.99, 9.82, \"alarm\"], [8.16, 9.24, \"alarm\"], [6.93, 6.7, \"alarm\"], [5.37, 7.99, \"alarm\"], [3.63, 5.94, \"ok\"], [6.79, 5.22, \"alarm\"], [2.84, 0.78, \"ok\"], [0.87, 3.56, \"ok\"], [5.8, 7.6, \"alarm\"], [7.14, 3.07, \"alarm\"], [9.29, 2.75, \"alarm\"], [7.16, 0.72, \"ok\"], [7.53, 6.7, \"alarm\"], [9.57, 8.97, \"alarm\"], [6.88, 8.38, \"alarm\"], [7.39, 6.11, \"alarm\"], [2.09, 5.17, \"ok\"], [8.96, 2.39, \"alarm\"], [9.75, 5.44, \"alarm\"], [3.93, 0.03, \"ok\"], [3.9, 1.78, \"ok\"], [6.53, 9.0, \"alarm\"], [9.1, 6.14, \"alarm\"], [3.87, 1.09, \"ok\"], [6.89, 5.53, \"alarm\"], [7.15, 3.74, \"alarm\"], [8.87, 2.13, \"alarm\"], [3.06, 2.68, \"ok\"], [6.22, 8.97, \"alarm\"], [1.67, 6.33, \"ok\"], [7.66, 2.25, \"ok\"], [0.63, 5.66, \"ok\"], [8.3, 8.79, \"alarm\"], [7.28, 4.17, \"alarm\"], [4.25, 5.29, \"ok\"], [9.05, 3.02, \"alarm\"], [2.81, 6.05, \"ok\"], [9.67, 1.87, \"alarm\"], [0.3, 1.16, \"ok\"], [5.63, 6.03, \"alarm\"], [1.84, 1.9, \"ok\"], [5.95, 6.46, \"alarm\"], [6.9, 7.29, \"alarm\"], [0.61, 4.85, \"ok\"], [8.35, 9.57, \"alarm\"], [3.18, 8.5, \"alarm\"], [6.45, 9.27, \"alarm\"], [2.29, 6.97, \"ok\"], [8.41, 4.77, \"alarm\"], [1.01, 1.94, \"ok\"], [1.57, 0.93, \"ok\"], [1.52, 6.16, \"ok\"], [1.05, 7.58, \"ok\"], [7.36, 8.16, \"alarm\"], [8.04, 7.13, \"alarm\"], [9.24, 9.57, \"alarm\"], [6.25, 9.65, \"alarm\"], [1.03, 1.03, \"ok\"], [0.64, 2.06, \"ok\"], [3.74, 4.59, \"ok\"], [6.81, 7.43, \"alarm\"], [0.84, 4.03, \"ok\"], [5.43, 3.75, \"ok\"]]";
ml sketch result "ht"

mentdb
{"count": 400, "nodes": 1, "leaves": 1, "depth": 0, "classes": ["alarm", "ok"], "prequential_accuracy": 0.9323308270676691}

ADWIN: positions of the detected changes

admin
ml sketch create "dr" "adwin";
ml sketch add "dr" "[11.0, 10.91, 8.93, 11.08, 7.44, 10.07, 10.36, 9.87, 10.1, 8.4, 9.86, 10.48, 9.07, 10.12, 9.82, 7.77, 9.39, 10.56, 8.46, 10.01, 10.32, 8.73, 9.74, 10.87, 10.7, 9.01, 10.99, 10.33, 10.11, 11.48, 10.71, 10.35, 10.94, 10.65, 11.24, 9.88, 8.24, 11.53, 10.79, 10.51, 10.04, 9.91, 10.09, 9.06, 10.76, 10.08, 9.05, 10.25, 9.67, 9.97, 10.53, 12.65, 8.94, 10.4, 10.7, 10.41, 11.46, 9.57, 9.67, 9.98, 12.64, 15.08, 12.77, 14.54, 15.18, 13.17, 13.82, 13.8, 15.69, 14.26, 12.26, 15.06, 15.02, 12.02, 11.3, 16.58, 14.08, 11.89, 15.04, 15.4, 11.64, 15.26, 14.7, 13.97, 13.3, 14.38, 13.19, 14.15, 13.74, 13.56, 14.54, 14.97, 14.88, 14.44, 15.66, 12.75, 14.06, 14.18, 15.2, 11.76]";
ml sketch result "dr"

mentdb
{"count": 100, "width": 40, "mean": 13.998249999999999, "drifts": 1, "drift_positions": [65]}

Closed itemsets of a stream of baskets

admin
ml sketch create "cl" "closed_itemsets";
ml sketch add "cl" "[[\"bread\", \"milk\"], [\"bread\", \"butter\"], [\"bread\", \"milk\", \"butter\"], [\"milk\"], [\"bread\", \"milk\"]]";
ml sketch result "cl"

mentdb
{"count": 5, "transactions": 5, "itemsets": [{"items": ["bread"], "support": 4, "frequency": 0.8}, {"items": ["milk"], "support": 4, "frequency": 0.8}, {"items": ["bread", "milk"], "support": 3, "frequency": 0.6}, {"items": ["bread", "butter"], "support": 2, "frequency": 0.4}]}

Recent frequent itemsets (older baskets count less)

admin
ml sketch create "rc" "recent_itemsets" "{\"min_support\": 0.3, \"half_life\": 3}";
ml sketch add "rc" "[[\"bread\", \"milk\"], [\"bread\", \"butter\"], [\"bread\", \"milk\", \"butter\"], [\"milk\"], [\"bread\", \"milk\"]]";
ml sketch result "rc"

mentdb
{"count": 5, "transactions": 5, "itemsets": [{"items": ["milk"], "recent_support": 0.7299059758563772}, {"items": ["bread"], "recent_support": 0.6414555842698305}, {"items": ["bread", "milk"], "recent_support": 0.4908763653695975}]}

Used in use cases (2 recipes)

IoT & sensor engineer › Quantiles and distinct devices of a telemetry stream
IoT & sensor engineer › Detect a change in a sensor stream

Often used with

ml sketch add · ml sketch create · ml sketch delete · ml sketch query · ml sketch cdf

ml sketch cdf <id> <x>

Description

tdigest: fraction of the values lower than or equal to x

Parameters

id: The sketch id - String - required
x: The value - Number - required

admin
ml sketch create "lat" "tdigest";
ml sketch add "lat" "[12, 15, 11, 19, 14, 120, 13, 16, 12, 18]";
ml sketch cdf "lat" 15

mentdb
0.6

Used in use cases (1 recipe)

IoT & sensor engineer › Quantiles and distinct devices of a telemetry stream

Often used with

ml sketch add · ml sketch create · ml sketch delete · ml sketch query · ml sketch result

ml sketch show

Description

The sketches in memory

admin
ml sketch create "visitors" "hll";
ml sketch show

mentdb
[{"id": "visitors", "type": "hll", "params": {}, "count": 0}]

ml sketch exist <id>

Description

Check if a sketch exists (1 or 0)

Parameters

id: The sketch id - String - required

admin
ml sketch exist "unknown_sketch"

mentdb
0

ml sketch delete <id>

Description

Delete a sketch

Parameters

id: The sketch id - String - required

admin
ml sketch create "visitors" "hll";
ml sketch delete "visitors"

mentdb
1

Used in use cases (2 recipes)

IoT & sensor engineer › Quantiles and distinct devices of a telemetry stream
IoT & sensor engineer › Detect a change in a sensor stream

Often used with

ml sketch add · ml sketch create · ml sketch query · ml sketch result · ml sketch cdf

ml sketch delete_all

Description

Delete all the sketches (returns the number of deleted sketches)

admin
ml sketch create "visitors" "hll";
ml sketch delete_all

mentdb
1