RAG approach - Generate predictions

To complete

1 Get labelled data

import duckdb
con = duckdb.connect(database=":memory:")

con.execute("INSTALL httpfs;")
con.execute("LOAD httpfs;")

# query_definition = f"SELECT * FROM read_parquet('s3://projet-formation/diffusion/funathon/2026/project2/generation_None_temp08.parquet')"
query_definition = f"SELECT * FROM read_parquet('https://minio.lab.sspcloud.fr/projet-formation/diffusion/funathon/2026/project2/generation_None_temp08.parquet')"
annotations = con.sql(query_definition).to_df()
annotations.head()

	code	name	label
0	01.11	Growing of cereals, other than rice, leguminou...	Pulses cultivation for market
1	01.11	Growing of cereals, other than rice, leguminou...	Legume crop production activities
2	01.11	Growing of cereals, other than rice, leguminou...	Broad bean farming operations
3	01.11	Growing of cereals, other than rice, leguminou...	Chickpea harvesting and processing
4	01.11	Growing of cereals, other than rice, leguminou...	Production of dried beans and peas

import os
from dotenv import load_dotenv
from qdrant_client import QdrantClient

load_dotenv()

client_qdrant = QdrantClient(
    url=os.environ["QDRANT_URL"],
    api_key=os.environ["QDRANT_API_KEY"],
    port=os.environ["QDRANT_API_PORT"]
)

collections = client_qdrant.get_collections()

NoteAttention

To complete

Tip Exercice 1: To complete

1. To complete.

print("Hello")

Hello

2. To complete.

print("Hello")

Hello

3. To complete.

Clue here

To complete.