Chapter 24

vector search

Notebookllm-zoomcamp-2026-vector45 cells

In [3]python · cell 1

python

print(123)

Output

In [7]python · cell 2

python

q1 = "I just discovered the course, can I still join?"
q2 = "I just found out about the program, can I still enroll?"

In [4]python · cell 3

python

from sentence_transformers import SentenceTransformer

model = SentenceTransformer('all-MiniLM-L6-v2')

Output

Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.

modules.json:   0%|          | 0.00/349 [00:00<?, ?B/s]

config_sentence_transformers.json:   0%|          | 0.00/116 [00:00<?, ?B/s]

README.md: 0.00B [00:00, ?B/s]

sentence_bert_config.json:   0%|          | 0.00/53.0 [00:00<?, ?B/s]

config.json:   0%|          | 0.00/612 [00:00<?, ?B/s]

model.safetensors:   0%|          | 0.00/90.9M [00:00<?, ?B/s]

Loading weights:   0%|          | 0/103 [00:00<?, ?it/s]

tokenizer_config.json:   0%|          | 0.00/350 [00:00<?, ?B/s]

vocab.txt: 0.00B [00:00, ?B/s]

tokenizer.json: 0.00B [00:00, ?B/s]

special_tokens_map.json:   0%|          | 0.00/112 [00:00<?, ?B/s]

config.json:   0%|          | 0.00/190 [00:00<?, ?B/s]

In [9]python · cell 4

python

v1 = model.encode(q1)

In [10]python · cell 5

python

v1.shape

Output

(384,)

In [12]python · cell 6

python

v2 = model.encode(q2)

In [13]python · cell 7

python

q1 = 'Can I still join the course after the start date?'
v1 = model.encode(q1)

In [14]python · cell 8

python

d  = "You don't need to register. You're accepted. You can also just start learning and submitting homework without registering."
dv = model.encode(d)

In [17]python · cell 9

python

v1.dot(dv)

Output

np.float32(0.32332397)

In [18]python · cell 10

python

q2 = 'How to install Docker on Windows?'
v2 = model.encode(q2)

In [19]python · cell 11

python

v2.dot(dv)

Output

np.float32(0.019730574)

In [20]python · cell 12

python

!wget https://raw.githubusercontent.com/DataTalksClub/llm-zoomcamp/main/01-agentic-rag/code/ingest.py

Output

--2026-05-18 10:31:25--  https://raw.githubusercontent.com/DataTalksClub/llm-zoomcamp/main/01-agentic-rag/code/ingest.py
Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.111.133, 185.199.108.133, 185.199.110.133, ...
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.111.133|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 738 [text/plain]
Saving to: ‘ingest.py’

ingest.py           100%[===================>]     738  --.-KB/s    in 0s      

2026-05-18 10:31:26 (41.5 MB/s) - ‘ingest.py’ saved [738/738]

In [21]python · cell 13

python

from ingest import load_faq_data

documents = load_faq_data()

In [23]python · cell 14

python

documents[10]

Output

{'id': '2b5ff70c77',
 'course': 'machine-learning-zoomcamp',
 'section': 'General Course-Related Questions',
 'question': 'Do I need to enroll in the course before submitting homework?',
 'answer': 'No enrollment is required to submit homework. Just log into the homework form when it opens. The Airtable registration you may see is only for announcements; actual submissions are made on the course platform forms and via your GitHub as specified in the homework guidelines.'}

In [ ]python · cell 15

python

texts = []

for doc in documents:
    text = doc['question'] + ' ' + doc['answer']
    texts.append(text)

In [26]python · cell 16

python

len(texts)

Output

In [ ]python · cell 17

python

documents[10]

Output

{'id': '2b5ff70c77',
 'course': 'machine-learning-zoomcamp',
 'section': 'General Course-Related Questions',
 'question': 'Do I need to enroll in the course before submitting homework?',
 'answer': 'No enrollment is required to submit homework. Just log into the homework form when it opens. The Airtable registration you may see is only for announcements; actual submissions are made on the course platform forms and via your GitHub as specified in the homework guidelines.'}

In [29]python · cell 18

python

from tqdm.auto import tqdm

batch_size = 50
vectors = []

for i in tqdm(range(0, len(texts), batch_size)):
    batch = texts[i:i + batch_size]
    batch_vectors = model.encode(batch)
    vectors.extend(batch_vectors)

len(vectors)

Output

  0%|          | 0/25 [00:00<?, ?it/s]

In [35]python · cell 19

python

scores = []

for i in range(len(vectors)):
    score = v1.dot(vectors[i])
    scores.append(score)

In [37]python · cell 20

python

import numpy as np
X = np.array(vectors)

In [40]python · cell 21

python

scores = X.dot(v1)

In [42]python · cell 22

python

idx = np.argmax(scores)
idx, scores[idx]

Output

(np.int64(553), np.float32(0.762941))

In [43]python · cell 23

python

documents[553]

Output

{'id': '3f1424af17',
 'course': 'data-engineering-zoomcamp',
 'section': 'General Course-Related Questions',
 'question': 'Course: Can I still join the course after the start date?',
 'answer': "Yes, even if you don't register, you're still eligible to submit the homework.\n\nBe aware, however, that there will be deadlines for turning in homeworks and the final projects. So don't leave everything for the last minute."}

In [51]python · cell 24

python

top5 = np.argsort(scores)[-5:]
top5 = top5[::-1]

In [52]python · cell 25

python

scores[top5]

Output

array([0.762941  , 0.7579372 , 0.7192131 , 0.6536311 , 0.56009984],
      dtype=float32)

In [53]python · cell 26

python

for idx in top5:
    print(scores[idx])
    print(documents[idx])
    print()

Output

0.762941
{'id': '3f1424af17', 'course': 'data-engineering-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'Course: Can I still join the course after the start date?', 'answer': "Yes, even if you don't register, you're still eligible to submit the homework.\n\nBe aware, however, that there will be deadlines for turning in homeworks and the final projects. So don't leave everything for the last minute."}

0.7579372
{'id': '2d8b16c2a0', 'course': 'mlops-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'Course - Can I still join the course after the start date?', 'answer': "Yes, even if you don't register, you're still eligible to submit the homeworks as long as the form is still open and accepting submissions.\n\nBe aware, however, that there will be deadlines for turning in the final projects. So don't leave everything to the last minute."}

0.7192131
{'id': '41aabbd7c5', 'course': 'machine-learning-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'The course has already started. Can I still join it?', 'answer': 'Yes, you can. Even though you missed the start date, you can register for the course. You won’t be able to submit some of the homeworks, but you can still take part in the course.\n\nIn order to get a certificate, you need to submit 2 out of 3 course projects and review 3 peers by the deadline. It means that if you join the course at the end of November and manage to work on two projects, you will still be eligible for a certificate.'}

0.6536311
{'id': '74eb249bbf', 'course': 'llm-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'I just discovered the course. Can I still join?', 'answer': 'Yes, but if you want to receive a certificate, you need to submit your project while we’re still accepting submissions.'}

0.56009984
{'id': '068529125b', 'course': 'data-engineering-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'Course - Can I follow the course after it finishes?', 'answer': 'Yes, we will keep all the materials available, so you can follow the course at your own pace after it finishes.\n\nYou can also continue reviewing the homeworks and prepare for the next cohort. You can also start working on your final capstone project.'}

In [59]python · cell 27

python

top5

Output

array([553, 955,  29, 472, 558])

In [60]python · cell 28

python

top5 = np.argsort(-scores)[:5]

In [61]python · cell 29

python

top5

Output

array([553, 955,  29, 472, 558])

In [62]python · cell 30

python

from minsearch import VectorSearch

vindex = VectorSearch(keyword_fields=['course'])
vindex.fit(X, documents)

Output

<minsearch.vector.VectorSearch at 0x70dabdde4590>

In [65]python · cell 31

python

vindex.search(v1, num_results=5, filter_dict={'course': 'llm-zoomcamp'})

Output

[{'id': '74eb249bbf',
  'course': 'llm-zoomcamp',
  'section': 'General Course-Related Questions',
  'question': 'I just discovered the course. Can I still join?',
  'answer': 'Yes, but if you want to receive a certificate, you need to submit your project while we’re still accepting submissions.'},
 {'id': 'bd31146b0e',
  'course': 'llm-zoomcamp',
  'section': 'General Course-Related Questions',
  'question': 'When will the course be offered next?',
  'answer': 'Summer 2025.'},
 {'id': '69d122f12e',
  'course': 'llm-zoomcamp',
  'section': 'General Course-Related Questions',
  'question': 'Certificate: Can I follow the course in a self-paced mode and get a certificate?',
  'answer': 'No, you can only get a certificate if you finish the course with a "live" cohort.\n\nWe don\'t award certificates for the self-paced mode. The reason is you need to peer-review 3 capstone(s) after submitting your project.\n\nYou can only peer-review projects at the time the course is running; after the form is closed and the peer-review list is compiled.'},
 {'id': '977bf7786c',
  'course': 'llm-zoomcamp',
  'section': 'General Course-Related Questions',
  'question': 'Course: I have registered for the LLM Zoomcamp. When can I expect to receive the confirmation email?',
  'answer': "You don't need it. You're accepted. You can also just start learning and submitting homework (while the form is open) without registering. It is not checked against any registered list. Registration is just to gauge interest before the start date."},
 {'id': '85384a18e5',
  'course': 'llm-zoomcamp',
  'section': 'Module 1: Introduction to LLMs and RAG',
  'question': 'OpenAI: Do I have to subscribe and pay for Open AI API for this course?',
  'answer': "No, you don't have to pay for this service in order to complete the course homeworks. You could use some of the free alternatives listed in the course GitHub.\n\n[llm-zoomcamp/01-intro/open-ai-alternatives.md at main · DataTalksClub/llm-zoomcamp (github.com)](https://github.com/DataTalksClub/llm-zoomcamp/blob/main/01-intro/open-ai-alternatives.md)"}]

In [66]python · cell 32

python

!wget https://raw.githubusercontent.com/DataTalksClub/llm-zoomcamp/main/01-agentic-rag/code/rag_helper.py

Output

--2026-05-18 10:56:54--  https://raw.githubusercontent.com/DataTalksClub/llm-zoomcamp/main/01-agentic-rag/code/rag_helper.py
Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.111.133, 185.199.110.133, ...
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 2134 (2.1K) [text/plain]
Saving to: ‘rag_helper.py’

rag_helper.py       100%[===================>]   2.08K  --.-KB/s    in 0s      

2026-05-18 10:56:54 (25.3 MB/s) - ‘rag_helper.py’ saved [2134/2134]

In [67]python · cell 33

python

from dotenv import load_dotenv
from openai import OpenAI

load_dotenv()
openai_client = OpenAI()

In [68]python · cell 34

python

from ingest import load_faq_data, build_index

documents = load_faq_data()
index = build_index(documents)

In [ ]python · cell 35

python

from rag_helper import RAGBase

assistant = RAGBase(
    index=index,
    llm_client=openai_client,
)

In [71]python · cell 36

python

query = 'I just found out about the program, can I still sign up?'
assistant.rag(query)

Output

'Yes — you can still sign up. If you want a certificate, make sure to submit your project while submissions are still open.'

In [73]python · cell 37

python

class RAGVector(RAGBase):

    def __init__(self, embedder, **kwargs):
        super().__init__(**kwargs)
        self.embedder = embedder

    def search(self, query, num_results=5):
        query_vector = self.embedder.encode(query)
        filter_dict = {'course': self.course}

        return self.index.search(
            query_vector,
            num_results=num_results,
            filter_dict=filter_dict
        )

In [ ]python · cell 38

python

vector_assistant = RAGVector(
    embedder=model,
    index=vindex,
    llm_client=openai_client,
)

In [75]python · cell 39

python

query = 'I just found out about the program, can I still sign up?'
vector_assistant.rag(query)

Output

'Yes, but if you want to receive a certificate, you need to submit your project while submissions are still open.'

In [76]python · cell 40

python

from sqlitesearch import VectorSearchIndex

vs_index = VectorSearchIndex(
    keyword_fields=['course'],
    mode='ivf',
    db_path='faq_vectors2.db'
)

In [77]python · cell 41

python

vs_index.fit(vectors, documents)

Output

<sqlitesearch.vector.index.VectorSearchIndex at 0x70dbbd1949e0>

In [80]python · cell 42

python

query = 'I just discovered the course. Can I still join it?'
query_vector = model.encode(query)

results = vs_index.search(query_vector, num_results=5)

In [81]python · cell 43

python

results = vs_index.search(
    query_vector,
    filter_dict={'course': 'llm-zoomcamp'},
    num_results=5
)

In [86]python · cell 44

python

vs_index.close()

In [ ]python · cell 45

python

--2026-05-18 10:31:25-- https://raw.githubusercontent.com/DataTalksClub/llm-zoomcamp/main/01-agentic-rag/code/ingest.py Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.111.133, 185.199.108.133, 185.199.110.133, ... Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.111.133|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 738 [text/plain] Saving to: ‘ingest.py’ ingest.py 100%[===================>] 738 --.-KB/s in 0s 2026-05-18 10:31:26 (41.5 MB/s) - ‘ingest.py’ saved [738/738]

{'id': '2b5ff70c77', 'course': 'machine-learning-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'Do I need to enroll in the course before submitting homework?', 'answer': 'No enrollment is required to submit homework. Just log into the homework form when it opens. The Airtable registration you may see is only for announcements; actual submissions are made on the course platform forms and via your GitHub as specified in the homework guidelines.'}

from tqdm.auto import tqdm batch_size = 50 vectors = [] for i in tqdm(range(0, len(texts), batch_size)): batch = texts[i:i + batch_size] batch_vectors = model.encode(batch) vectors.extend(batch_vectors) len(vectors)

{'id': '3f1424af17', 'course': 'data-engineering-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'Course: Can I still join the course after the start date?', 'answer': "Yes, even if you don't register, you're still eligible to submit the homework.\n\nBe aware, however, that there will be deadlines for turning in homeworks and the final projects. So don't leave everything for the last minute."}

0.762941 {'id': '3f1424af17', 'course': 'data-engineering-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'Course: Can I still join the course after the start date?', 'answer': "Yes, even if you don't register, you're still eligible to submit the homework.\n\nBe aware, however, that there will be deadlines for turning in homeworks and the final projects. So don't leave everything for the last minute."} 0.7579372 {'id': '2d8b16c2a0', 'course': 'mlops-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'Course - Can I still join the course after the start date?', 'answer': "Yes, even if you don't register, you're still eligible to submit the homeworks as long as the form is still open and accepting submissions.\n\nBe aware, however, that there will be deadlines for turning in the final projects. So don't leave everything to the last minute."} 0.7192131 {'id': '41aabbd7c5', 'course': 'machine-learning-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'The course has already started. Can I still join it?', 'answer': 'Yes, you can. Even though you missed the start date, you can register for the course. You won’t be able to submit some of the homeworks, but you can still take part in the course.\n\nIn order to get a certificate, you need to submit 2 out of 3 course projects and review 3 peers by the deadline. It means that if you join the course at the end of November and manage to work on two projects, you will still be eligible for a certificate.'} 0.6536311 {'id': '74eb249bbf', 'course': 'llm-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'I just discovered the course. Can I still join?', 'answer': 'Yes, but if you want to receive a certificate, you need to submit your project while we’re still accepting submissions.'} 0.56009984 {'id': '068529125b', 'course': 'data-engineering-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'Course - Can I follow the course after it finishes?', 'answer': 'Yes, we will keep all the materials available, so you can follow the course at your own pace after it finishes.\n\nYou can also continue reviewing the homeworks and prepare for the next cohort. You can also start working on your final capstone project.'}

[{'id': '74eb249bbf', 'course': 'llm-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'I just discovered the course. Can I still join?', 'answer': 'Yes, but if you want to receive a certificate, you need to submit your project while we’re still accepting submissions.'}, {'id': 'bd31146b0e', 'course': 'llm-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'When will the course be offered next?', 'answer': 'Summer 2025.'}, {'id': '69d122f12e', 'course': 'llm-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'Certificate: Can I follow the course in a self-paced mode and get a certificate?', 'answer': 'No, you can only get a certificate if you finish the course with a "live" cohort.\n\nWe don\'t award certificates for the self-paced mode. The reason is you need to peer-review 3 capstone(s) after submitting your project.\n\nYou can only peer-review projects at the time the course is running; after the form is closed and the peer-review list is compiled.'}, {'id': '977bf7786c', 'course': 'llm-zoomcamp', 'section': 'General Course-Related Questions', 'question': 'Course: I have registered for the LLM Zoomcamp. When can I expect to receive the confirmation email?', 'answer': "You don't need it. You're accepted. You can also just start learning and submitting homework (while the form is open) without registering. It is not checked against any registered list. Registration is just to gauge interest before the start date."}, {'id': '85384a18e5', 'course': 'llm-zoomcamp', 'section': 'Module 1: Introduction to LLMs and RAG', 'question': 'OpenAI: Do I have to subscribe and pay for Open AI API for this course?', 'answer': "No, you don't have to pay for this service in order to complete the course homeworks. You could use some of the free alternatives listed in the course GitHub.\n\n[llm-zoomcamp/01-intro/open-ai-alternatives.md at main · DataTalksClub/llm-zoomcamp (github.com)](https://github.com/DataTalksClub/llm-zoomcamp/blob/main/01-intro/open-ai-alternatives.md)"}]

--2026-05-18 10:56:54-- https://raw.githubusercontent.com/DataTalksClub/llm-zoomcamp/main/01-agentic-rag/code/rag_helper.py Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.111.133, 185.199.110.133, ... Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 2134 (2.1K) [text/plain] Saving to: ‘rag_helper.py’ rag_helper.py 100%[===================>] 2.08K --.-KB/s in 0s 2026-05-18 10:56:54 (25.3 MB/s) - ‘rag_helper.py’ saved [2134/2134]

class RAGVector(RAGBase): def __init__(self, embedder, **kwargs): super().__init__(**kwargs) self.embedder = embedder def search(self, query, num_results=5): query_vector = self.embedder.encode(query) filter_dict = {'course': self.course} return self.index.search( query_vector, num_results=num_results, filter_dict=filter_dict )