Scribd Logo
Upload

Welcome to Scribd!

  • 5 views

    Vector Embedding

    Uploaded by

    riteshtarun6

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Vector Embedding

    Uploaded by

    riteshtarun6
    5 views8 pages

    Original Title

    vector_Embedding

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    5 views8 pages

    Vector Embedding

    Uploaded by

    riteshtarun6

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 8

    Understand Basis of

    building Ai powered apps

    What is
    vector
    Embedding?
    M Tanusri 01
    Vector embeddings are numerical
    representations of words or
    objects in a continuous vector
    space.

    super fun

    im happy
    angry Love it

    Enjoyy

    Can’t tollerate

    Messenger

    Networking

    Facebook

    Instagram
    Capturing the semantic
    meaning, allowing machines
    to understand and process
    language more effectively.
    For example a word
    "Honey" has a vector representation
    [1.5, -0.4, 7.2, 19.6, 20.2]

    Similarly we have words and phrases


    that can be represented as

    "Samosa"- [8.1, -3.9, 7.8, 1.6, 80.2]


    "I am happy"- [2.9, -1.4, 0.2, 0.6, 2.2]
    "Redbull gives you wiiings"- [7.9, -4.0, 3.2, 14.2, 21.2]

    Now these representations are called vector embeddings


    Wondering how do we get
    these values ?
    Lets see how
    vector
    embeddings
    are created ->
    Embeddings are typically generated using
    pretrained machine learning models such as
    OpenAI, Cohere, BERT, and many others.

    While training these models on large corpora


    of text, they learn to represent words in a
    vector space based on their relationships
    with other words.

    These vectors are not just numbers


    assigned to words, but also represent their
    semantic nature.

    Example-
    "Happy" is similar to "enjoyed" are close to
    each other where as "Sad" is opposite in its
    meaning thus its far away in the space.
    distance(d)
    Happy [1.1, 2.3, 3.4, 01]
    Angry [7.7, 8.8, 3.4, 9.9]
    Enjoyed [1.1, 2.3, 3.4, 02]
    What are its
    applications and
    why is it
    important?
    Vector embeddings are basic
    fundamentals of
    RAG (retrieval augmented generation)
    which is indeed a backbone of building
    any Ai applications
    Vector embeddings are used in
    various fields,

    > Natural Language Processing,


    > Recommendation Systems,
    > Image Recognition,
    > Sentiment Analysis,
    > To enhance search engines,
    > Personalize Recommendations,
    > identify objects in images, and
    understand opinions in text.
    Knowing this concept we
    could build cool projects!
    Up next we will learn about
    ✌️
    “RAG” in detail

    Follow for more content like


    this....

    @Tanusri

    You might also like