I didn’t understand this part at all. We just ended simple hashing by remaining of deviding. So, what are this planes at all? As I understand we dedicate some area of values of the same category by this planes. And depending of signs of np.dot(P, v.T) we understand is vector v part of this area. Wh…

Hi @someone555777 [image] someone555777: I didn’t understand this part at all. We just ended simple hashing by remaining of dividing. So, what are this planes at all? As I understand we dedicate some area of values of the same category by this planes. We use use planes to divide up the space…

emm, ok, thank you for an answer. But should I understand this at all? As I understand that it is only one of techniques to cluster words with similar meaning? And not very often used?

[image] someone555777: But should I understand this at all? :slight_smile: You will need it later in the course. And the general answer depends on your goals in learning (from this Course and overall). If, for example, you’re just poking around and plan using prompt engineering or you’re som…

and one more question. If I understand correct, we find the side of plane of the vector, not only any of dots. So, how is it visualizated at all? As I understand vector can cross the plane in most part of cases. Should we orient by side on which most part of this vector is?

and one more question, do I understand correct, that hash_multi_plane_matrix(random_planes_matrix, v, num_planes) from lab “Hash tables” that takes 3 planes can output only one of 1-7 digits? That are potential places in plot that are formed by crossing of this 3 lines?

[image] someone555777: So, how is it visualizated at all? [studentui2] Each dot is a vector (an instance of data, green-positive, red-negative) Each colored line is a plane (random plane dividing data) After this division you limit the search of the nearest points (vectors) in between the …

here are dots = vectors, for example. Why any of this dots were uncotegorized by the way? [изображение]

and how do I get this separation planes by the way? Is it something like gradient decent?

By random, they are not learned (no gradient descent).

emm, how is it? why do we need them at all if they are random? do they help much in clusterisation? And what about my question about screen from the course?

Why do we need Multiple Planes?

Course Q&A Natural Language Processing NLP with Classification and Vector Spaces

arvyzukai June 6, 2023, 3:15pm 7

studentui2

Each dot is a vector (an instance of data, green-positive, red-negative)
Each colored line is a plane (random plane dividing data)

After this division you limit the search of the nearest points (vectors) in between the colored lines.

Topic		Replies	Views
Question on locality-sensitive hashing NLP with Classification and Vector Spaces week-module-4	17	532	January 23, 2024
Question about The approximate nearest neighbor algorithm NLP with Classification and Vector Spaces week-module-4	6	558	April 21, 2023
Reading: Multiple Planes / Planes crossing at the origin NLP with Classification and Vector Spaces week-module-4	6	556	July 8, 2022
C1_W4_Assignment - Choosing number of planes NLP with Classification and Vector Spaces week-module-4	1	314	November 16, 2023
Exercise 09 - hash tables, vector multiplication NLP with Classification and Vector Spaces week-module-4	1	573	June 7, 2022

Why do we need Multiple Planes?

Related topics