$
docker run -d --name redis-stack -p 6379:6379 -p 8001:8001 redis/redis-stack:latest

$
!pip install redisvl==0.1.0
>
!pip install cohere==4.45
>
!pip install jsonlines

1
from redis import Redis
2
from redisvl.index import SearchIndex
3
from redisvl.schema import IndexSchema
4
from redisvl.utils.vectorize import CohereTextVectorizer
5
from redisvl.query import VectorQuery
6
from redisvl.query.filter import Tag, Text, Num
7
import jsonlines

1
version: "0.1.0"
2
index:
3
  name: semantic_search_demo
4
  prefix: rvl
5
  storage_type: hash
6
7
fields:
8
  - name: url
9
    type: text
10
  - name: title
11
    type: tag
12
  - name: text
13
    type: text
14
  - name: wiki_id
15
    type: numeric
16
  - name: paragraph_id
17
    type: numeric
18
  - name: id
19
    type: numeric
20
  - name: views
21
    type: numeric
22
  - name: langs
23
    type: numeric
24
  - name: embedding
25
    type: vector
26
    attrs:
27
      algorithm: flat
28
      dims: 1024
29
      distance_metric: cosine
30
      datatype: float32

1
# create a vectorizer
2
api_key='{Insert your cohere API Key}'
3
4
cohere_vectorizer = CohereTextVectorizer(
5
    model="embed-english-v3.0",
6
    api_config={"api_key": api_key},
7
)

1
# construct a search index from the schema - this schema is called "semantic_search_demo"
2
schema = IndexSchema.from_yaml("./schema.yaml")
3
client = Redis.from_url("redis://localhost:6379")
4
index = SearchIndex(schema, client)
5
6
# create the index (no data yet)
7
index.create(overwrite=True)

1
!rvl index listall

15:39:22 [RedisVL] INFO   Indices:
15:39:22 [RedisVL] INFO   1. semantic_search_demo

1
!rvl index info -i semantic_search_demo

1
# read in your documents
2
jsonl_file_path='data/redis_guide_data.jsonl'
3
4
corpus=[]
5
text_to_embed=[]
6
7
with jsonlines.open(jsonl_file_path, mode='r') as reader:
8
    for line in reader:
9
        corpus.append(line)
10
        # we want to store the embeddings of the field called `text`
11
        text_to_embed.append(line['text'])
12
13
# call embed_many which returns an array
14
# hash data structures get serialized as a string and thus we store the embeddings in hashes as a byte string (handled by numpy)
15
res=cohere_vectorizer.embed_many(text_to_embed, input_type='search_document', as_buffer=True)

1
# contruct the data payload to be uploaded to your index
2
data = [{"url": row['url'],
3
         "title": row['title'],
4
         "text": row['text'],
5
         "wiki_id": row['wiki_id'],
6
         "paragraph_id": row['paragraph_id'],
7
         "id":row['id'],
8
         "views":row['views'],
9
         "langs":row['langs'],
10
         "embedding":v}
11
        for row, v in zip(corpus, res)]
12
13
# load the data into your index
14
index.load(data)

1
# use the Cohere vectorizer again to create a query embedding
2
query_embedding = cohere_vectorizer.embed("What did Microsoft release in 2015?", input_type='search_query',as_buffer=True)
3
4
5
query = VectorQuery(
6
    vector=query_embedding,
7
    vector_field_name="embedding",
8
    return_fields=["url","wiki_id","paragraph_id","id","views","langs","title","text",],
9
    num_results=5
10
)
11
12
results = index.query(query)
13
14
for doc in results:
15
    print(f"Title:{doc['title']}\nText:{doc['text']}\nDistance {doc['vector_distance']}\n\n")

1
# Initialize a tag filter
2
tag_filter = Tag("title") == "Microsoft Office"
3
4
# set the tag filter on our existing query
5
query.set_filter(tag_filter)
6
7
results = index.query(query)
8
9
for doc in results:
10
    print(f"Title:{doc['title']}\nText:{doc['text']}\nDistance {doc['vector_distance']}\n")

1
# define a tag match on the title, text match on the text field, and numeric filter on the views field
2
filter_data=(Tag('title')=='Elizabeth II') & (Text("text")% "born") & (Num("views")>4500)
3
4
query_embedding = co.embed("When was she born?", input_type='search_query',as_buffer=True)
5
6
# reinitialize the query with the filter expression
7
query = VectorQuery(
8
    vector=query_embedding,
9
    vector_field_name="embedding",
10
    return_fields=["url","wiki_id","paragraph_id","id","views","langs","title","text",],
11
    num_results=5,
12
    filter_expression=filter_data
13
)
14
15
results = index.query(query)
16
print(results)
17
18
for doc in results:
19
    print(f"Title:{doc['title']}\nText:{doc['text']}\nDistance {doc['vector_distance']}\nView {doc['views']}")