redisml

Machine Learning Model Server

RedisLabsModules
306
44
C

Notice RedisML is planned to be replaced by RedisAI, adding support for deep learning.

GitHub issues
CircleCI
DockerHub

docs/images/logo.png

Machine Learning Model Server on Redis

Overview

RedisML is a Redis module that implements several machine learning models as Redis data types.

The stored models are fully operational and support the prediction/evaluation process.

RedisML is a turnkey solution for using trained models in a production environment. Load ML models from any platform, immediately ready to serve.

See Full Documentation at http://redisml.io

Primary Features

  • Decision Tree ensembles (random forests) classification and regression
  • Linear regression
  • Logistic regression
  • Matrix operations

Building and Running

  • Build a Redis server with support for modules (currently available from the unstable branch).

  • You’ll also need a BLAS library such as ATLAS. To install ATLAS:

    • Ubuntu:
    sudo apt-get install libatlas-base-dev
    • CentOS/RHEL/Fedora:
    sudo yum install -y atlas-devel atlas-static
ln -s /usr/lib64/atlas/libatlas.a /usr/lib64/libatlas.a 
ln -s /usr/lib64/atlas/libtatlas.so /usr/lib64/libcblas.a

  • Build the RedisML module:

    git clone https://github.com/RedisLabsModules/redisml.git
cd redisml/src
make

  • To load the module, start Redis with the --loadmodule /path/to/redisml/src/redis-ml.so option, add it as a directive to the configuration file or send a MODULE LOAD command.

Redis ML Commands

Decision tree ensembles

Example of use

The following code creates a random forest under the key myforest that consists of three trees with IDs ranging from 0 to 2, where each consists of a single numeric splitter and its predicate values. Afterwards, the forest is used to classify two inputs and yield their predictions.

redis> ML.FOREST.ADD myforest 0 . NUMERIC 1 0.1 .l LEAF 1 .r LEAF 0
OK
redis> ML.FOREST.ADD myforest 1 . NUMERIC 1 0.1 .l LEAF 1 .r LEAF 0
OK
redis> ML.FOREST.ADD myforest 2 . NUMERIC 1 0.1 .l LEAF 0 .r LEAF 1
OK
redis> ML.FOREST.RUN myforest 1:0.01 CLASSIFICATION
"1"
redis> ML.FOREST.RUN myforest 1:0.2 CLASSIFICATION
"0"

ML.FOREST.ADD

Available since 1.0.0.
Time complexity: O(M*log(N)) where N is the tree’s depth and M is the number of nodes added

Syntax

ML.FOREST.ADD key tree path ((NUMERIC|CATEGORIC) attr val | LEAF val [STATS]) [...]

Description

Add nodes to a tree in the forest.
This command adds one or more nodes to the tree in the forest that’s stored under key. Trees are identified by numeric IDs, treeid, that must begin at 0 and be incremented by exactly 1 for each new tree.

Each of the nodes is described by its path and definition. The path argument is the path from the tree’s root to the node. A valid path always starts with the period character (.), which denotes the root. Optionally, the root may be followed by left or right branches, denoted by the characters l and r, respectively. For example, the path “.lr” refers to the right child of the root’s left child.

A node in the decision tree can either be a splitter or a terminal leaf. Splitter nodes are either numerical or categorical, and are added using the NUMERIC or CATEGORIC keywords. Splitter nodes also require specifying the examined attribute (attr) as well as the value (val) used in the comparison made during the branching decision. val is expected to be a double-precision floating point value for numerical splitters, and a string for categorical splitter nodes.

The leaves are created with the LEAF keyword and only require specifying their double-precision floating point value (val).

Return value:

Simple string reply

ML.FOREST.RUN

Available since 1.0.0.
Time complexity: O(M*log(N)) where N is the depth of the trees and M is the number of trees in the forest

Syntax

ML.FOREST.RUN key sample (CLASSIFICATION|REGRESSION)

Description

Predicts the classified (discrete) or regressed (continuous) value of a sample using the forest.
The forest that’s stored in key is used for generating the predicted value for the sample. The sample is given as a string that is a vector of attribute-value pairs in the format of attr:val. For example, the sample “gender:male” has a single attribute, gender, whose value is male. A sample may have multiple such attribute-value pairs, and these must be comma-separated (,) in the string vector. For example, a sample of a 25-years-old male is expressed as “gender:male,age:25”.

Return value:

Bulk string reply: the predicted value of the sample

Linear regression

Example of use

The first line of the example shows how a linear regression predictor is set to the key named linear. The predictor has an intercept of 2 and its coefficients are 3, 4 and 5. Once the predictor is ready, it is used to predict the result given the independent variables’ values (features) of 1, 1 and 1.

redis> ML.LINREG.SET linear 2 3 4 5
OK
redis> ML.LINREG.PREDICT linear 1 1 1
"14"

ML.LINREG.SET

Available since 1.0.0.
Time complexity: O(N) where N is the number of coefficients

Syntax

ML.LINREG.SET key intercept coefficient [...]

Description

Sets a linear regression predictor.
This command creates or updates the linear regression predictor that’s stored in key. The predictor’s intercept is specified by intercept, followed by one or more coefficient arguments of the independent variables.

Return value:

Simple string reply

ML.LINREG.PREDICT

Available since 1.0.0.
Time complexity: O(N) where N is the number of features

Syntax

ML.LINREG.PREDICT key feature [...]

Description

Predicts the result for a set of features.
The linear regression predictor stored in key is used for predicting the result based on one or more features that are given by the feature argument(s).

Return value:

Bulk string reply: the predicted result for the feature set

Logistic regression

Example of use

In this example, the first line shows how a logistic regression predictor is set to the key named logistic. The predictor has an intercept of 0 and its coefficients are 2 and 2. Once the predictor is ready, it is used to predict the result given the independent variables’ values (features) of -3 and 1.

redis> ML.LOGREG.SET logistic 0 2 2
OK
redis> ML.LOGREG.PREDICT logistic -3 1
"0.017986209962091559"

ML.LOGREG.SET

Available since 1.0.0.
Time complexity: O(N) where N is the number of coefficients

Syntax

ML.LOGREG.SET key intercept coefficient [...]

Description

Sets a linear regression predictor.
This command sets or updates the logistic regression predictor that’s stored in key. The predictor’s intercept is specified by intercept, followed by one or more coefficient arguments of the independent variables.

Return value:

Simple string reply

ML.LOGREG.PREDICT

Available since 1.0.0.
Time complexity: O(N) where N is the number of features

Syntax

ML.LOGREG.PREDICT key feature [...]

Description

Predicts the result for a set of features.
The logistic regression predictor stored in key is used for predicting the result based on one or more features that are given by the feature argument(s).

Return value:

Bulk string reply: the predicted result for the feature set

Matrix operations

Example of use

The following example shows how to set two matrices, a and b, multiply them, and store the result in the matrix ab. Lastly, the contents of ab are fetched.

redis> ML.MATRIX.SET a 2 3 1 2 5 3 4 6
OK
redis> ML.MATRIX.SET b 3 2 1 2 3 4 7 1
OK
redis> ML.MATRIX.MULTIPLY a b ab
OK
redis> ML.MATRIX.GET ab
1) (integer) 2
2) (integer) 2
3) "42"
4) "15"
5) "57"
6) "28"

ML.MATRIX.SET

Available since 1.0.0.
Time complexity: O(N*M) where N is the number of rows and M is the number of columns

Syntax

ML.MATRIX.SET key n m entry11 .. entrynm

Description

Sets a matrix.
Sets key to store a matrix of n rows,m columns and double-precision float entries ranging from entry11 to entrynm.

Return value:

Simple string reply

ML.MATRIX.GET

Available since 1.0.0.
Time complexity: O(N*M) where N is the number of rows and M is the number of columns

Syntax

ML.MATRIX.GET key

Description

Get a matrix.
Returns the matrix’s dimensions and entries.

Return value:

The first two elements in the returned array are the matrix’s rows and columns, respectively, followed by the entries.

ML.MATRIX.ADD

Available since 1.0.0.
Time complexity: O(N*M) where N is the number of rows and M is the number of columns

Syntax

ML.MATRIX.ADD matrix1 matrix2 sum

Description

Adds matrices.
The result of adding the two matrices stored in matrix1 and matrix2 is set in sum.

Return value:

Simple string reply

ML.MATRIX.MULTIPLY

Available since 1.0.0.
Time complexity: O(N*M*P) where N and M are numbers of rows and columns in matrix1, and P is the number of columns in matrix2

Syntax

ML.MATRIX.MULTIPLY matrix1 matrix2 product

Description

Multiplies matrices.
The result of multiplying the two matrices stored in matrix1 and matrix2 is set in product.

Return value:

Simple string reply

ML.MATRIX.SCALE

Available since 1.0.0.
Time complexity: O(N*M) where N is the number of rows and M is the number of columns

Syntax

ML.MATRIX.SCALE key scalar

Description

Scales a matrix.
Updates the entries of the matrix stored in key by multiplying them with scalar.

Return value:

Simple string reply

K-means

Example of use

Setting up a K-means model in key k with 2 clusters and 3 dimensions. The cluster centers are 1, 1, 2 and 2, 5, 4:

redis> ML.KMEANS.SET k 2 3 1 1 2 2 5 4
OK

Predicting the cluster of feature vector 1, 3, 5:

redis> ML.KMEANS.predict k 1 3 5 
(integer) 1

ML.KMEANS.SET

Available since 1.0.0.
Time complexity: O(N) where N is the number of coefficients

Syntax

ML.KMEANS.SET key k dimensions centers [...]

Description

Create/update a K-means model.
This command creates or updates the K-means model that’s stored in key. The number of classes is specified by k, the number of features is set by dimensions .

Return value:

Simple string reply

ML.KMEANS.PREDICT

Available since 1.0.0.
Time complexity: O(N) where N is the number of features

Syntax

ML.KMEANS.PREDICT key feature [...]

Description

Predicts the result for a set of features.
The K-means model stored in key is used for predicting the result based on one or more features that are given by the feature argument(s).

Return value:

Integer reply: the predicted result for the feature set

Contributing

Issue reports, pull and feature requests are welcome.

License

Redis Source Available License Agreement - see LICENSE