LLM Zoomcamp 2025: Module 1 Homework

1 Problem 1

Run Elastic Search 8.17.6

docker run -it \
  --rm \
  --name elasticsearch \
  -m 4GB \
  -p 9200:9200 \
  -p 9300:9300 \
  -e "discovery.type=single-node" \
  -e "xpack.security.enabled=false" \
  docker.elastic.co/elasticsearch/elasticsearch:8.17.6

curl localhost:9200

The version.build_hash value is "dbcbbbd0bc4924cfeb28929dc05d82d662c527b7"

2 Getting the data

import requests 

docs_url = 'https://github.com/DataTalksClub/llm-zoomcamp/blob/main/01-intro/documents.json?raw=1'
docs_response = requests.get(docs_url)
documents_raw = docs_response.json()

documents = []

for course in documents_raw:
    course_name = course['course']

    for doc in course['documents']:
        doc['course'] = course_name
        documents.append(doc)

3 Problem 2

from elasticsearch import Elasticsearch
from tqdm.auto import tqdm

es_client = Elasticsearch('http://localhost:9200')
index_settings = {
    "settings": {
        "number_of_shards": 1,
        "number_of_replicas": 0
    },
    "mappings": {
        "properties": {
            "text": {"type": "text"},
            "section": {"type": "text"},
            "question": {"type": "text"},
            "course": {"type": "keyword"} 
        }
    }
}

index_name = "course-questions"

# Check if the index already exists
if es_client.indices.exists(index=index_name):
    es_client.indices.delete(index=index_name)

# Create the index
es_client.indices.create(index=index_name, body=index_settings)

# Build the index
for doc in tqdm(documents):
    es_client.index(index=index_name, document=doc)

The function used for adding data to elastic is index

4 Problem 3

def elastic_search_3(query):
    search_query = {
        "size": 5,
        "query": {
            "bool": {
                "must": {
                    "multi_match": {
                        "query": query,
                        "fields": ["question^4", "text"],
                        "type": "best_fields"
                    }
                },
            }
        }
    }

    response = es_client.search(index=index_name, body=search_query)
    
    result_docs = []
    
    for hit in response['hits']['hits']:
        result_docs.append(hit['_source'])

    if response['hits']['total']['value'] > 0:
        top_score = response['hits']['hits'][0]['_score']
    else:
        top_score = 0
    
    return result_docs, top_score

docs_3, top_score_3 = elastic_search_3("How do execute a command on a Kubernetes pod?")
print(f"Top score: {top_score_3}")

Top score: 42.848103

5 Problem 4

def elastic_search_4(query):
    search_query = {
        "size": 3,
        "query": {
            "bool": {
                "must": {
                    "multi_match": {
                        "query": query,
                        "fields": ["question^4", "text"],
                        "type": "best_fields"
                    }
                },
                "filter": {
                    "term": {
                        "course": "machine-learning-zoomcamp"
                    }
                }                
            }
        }
    }

    response = es_client.search(index=index_name, body=search_query)
    
    result_docs = []
    
    for hit in response['hits']['hits']:
        result_docs.append(hit['_source'])

    if response['hits']['total']['value'] > 0:
        top_score = response['hits']['hits'][0]['_score']
    else:
        top_score = 0
    
    return result_docs, top_score

docs_4, top_score_4 = elastic_search_4("How do copy a file to a Docker container?")
print(f"The 3rd question returned is: {docs_4[2]['question']}")

The 3rd question returned is: How do I copy files from a different folder into docker container’s working directory?

6 Problem 5

def build_prompt(query, search_results):
    prompt_template = """
You're a course teaching assistant. Answer the QUESTION based on the CONTEXT from the FAQ database.
Use only the facts from the CONTEXT when answering the QUESTION.

QUESTION: {question}

CONTEXT:
{context}
""".strip()

    context_template = """
Q: {question}
A: {text}
""".strip()

    # Build context using the new context_template
    context_entries = []
    for doc in search_results:
        entry = context_template.format(question=doc["question"], text=doc["text"])
        context_entries.append(entry)
    
    context = "\n\n".join(context_entries)

    # Format the final prompt
    prompt = prompt_template.format(question=query, context=context).strip()
    return prompt

prompt = build_prompt("How do I execute a command in a running docker container?", docs_4)
print(f"The length of the resulting prompt is: {len(prompt)}")

The length of the resulting prompt is: 1462

7 Problem 5

import tiktoken
encoding = tiktoken.encoding_for_model("gpt-4o")
num_tokens = len(encoding.encode(prompt))
print(f"The length of tokens is: {num_tokens}")

The length of tokens is: 322