Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from http://reference.wolfram.com/language/ref/CreateVectorDatabase.html below:

Create a vector database—Wolfram Documentation

We deliver solutions for the AI era—combining symbolic computation, data-driven insights and deep technical expertise

• Data & Computational Intelligence
• Model-Based Design
• Algorithm Development
• Wolfram|Alpha for Business
• Blockchain Technology
• Education Technology
• Quantum Computation

WolframConsulting.com

• CreateVectorDatabase initializes a new vector database to store and manage high-dimensional data for efficient search and retrieval.
• Typical applications of vector databases include recommendation systems, image and text retrieval, and similarity searches in large datasets.
• Possible values for data include:
• {vec₁,…} a list of vectors {vec₁val₁,…} a list of vectors and associated values {vec₁,…}{val₁,…} a rule between vectors and values
• A list of scalars is interpreted as a list of one-dimensional vectors. »
• Accepted forms of val_i include:
• "string" string labels <|"tag₁"v₁,…|> an association of tags and metadata values
• The database name must be a string.
• The following options can be specified:
• Possible values for DistanceFunction include EuclideanDistance, SquaredEuclideanDistance, CosineDistance, JaccardDissimilarity and HammingDistance.
• Possible settings for WorkingPrecision include:
• "Integer8" signed 8-bit integers from through 127 "Real32" single-precision real (32 bit) "Real64" double-precision real (64 bit)

Create an empty VectorDatabaseObject:

Create a database with a specific name and initialize it with a list of vectors:

Search the database by similarity:

Create an empty vector database:

Create a database with a specific name:

Initialize a new vector database with a list of vectors:

Initialize the database with both vectors and metadata:

Specify the metadata as a separate list:

Specify the metadata with tags as an Association:

Specify data and metadata separately:

Specify a custom distance function for the database:

By default, EuclideanDistance is used:

Only vectors can be stored in the database; specify a FeatureExtractor that can extract image features:

Search for the closest image to a given one:

Specify a custom location to store the database:

Retrieve the location:

By default, the database is stored in a local object:

Store the vector database in a file:

Retrieve the location:

Recreate the database from the file reference:

Specify a custom working precision for the distance computation:

By default, the precision is inferred from the input data:

Specify a precision for an empty database:

Newly added vectors are clipped and/or rounded to fit into the given precision:

An empty database is created without an explicit working precision:

The value is resolved when data is added to the database:

A list of scalars is interpreted as a list of one-dimensional vectors:

Use an outer list to indicate that the input is a single vector:

Only one-dimensional arrays can be stored in the database:

All arrays in the database should have the same size:

The database name is unique:

A new database with the same name specification will be created with an incrementing suffix:

Use DeleteObject to remove an incorrectly generated VectorDatabaseObject:

Now the name is available again:

Alternatively, use the option OverwriteTargetTrue to overwrite the old database:

OverwriteTarget -> False will issue an error rather than adding an incrementing number:

Create a database at a specific location:

A database with the same name in a new location will be created with an incrementing suffix:

Delete the original database to free the name:

Now the name is available again:

RetroSearch is an open source project built by @garambo | Open a GitHub Issue

Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo

HTML: 3.2 | Encoding: UTF-8 | Version: 0.7.4