Most similar Text / Recommend items

This task calculates the text similarity of multiple item to a query.

Description

The task offers calculates the (cosine) similarity between a query text and item texts. It allows to enter a list of item texts and a query that is compared with the item texts.

Applications

• Recommendation: List the most similar items.

• Sentiment Analysis: Understand sentiments by comparing text similarities.

• Text Classification: Aid in categorizing and organizing texts efficiently.

How to use the task

Input the query, the items and the embedding model (that defines how to compare the texts) you want to use.
Click the "Run Task" button.
Instantly, you'll receive a similarity score that quantifies the items and their similarity to the query ordered by their similarity to the query.

Privacy

Please do not use any sensitive input.

[1.0.0]:

First version

Version	AI Model	Created	Link
1.0.0	openai	30.07.2023

API

The REST API allows you to call the tool with the same costs as when running the tool. Please generate an Personal access token before using the REST API.

Parameters

query (Query): The query that should be compared to the other items
items (Items): The items that should be compared to the query
model (Embedding model): Defines the embedding model

Call the REST API by cURL

curl -v -H "Authorization: Bearer PERSONAL_ACCESS_TOKEN" https://api.anysolve.ai/rest/v1/intern-embeddings-text-recommendation/1.0.0?query=States%20of%20the%20USA&items=Texas&model=text-embedding-ada-002

Install the package with pip
python3 -m pip install anysolve

Run in python3

import os
from anysolve import AnySolve

anysolve_token = os.environ.get('ANYSOLVE_PERSONAL_ACCESS_TOKEN') # Resolve your personal access token here
client = AnySolve(anysolve_token)
res = client.run('intern-embeddings-text-recommendation','1.0.0', {'query': 'States of the USA', 'items': 'Texas', 'model': 'text-embedding-ada-002'})

print(res)

Coming soon: Within AnySolve ChatComplete prompts you can use the following command to execute the task:
/run('intern-embeddings-text-recommendation','1.0.0', query='States of the USA', items='Texas', model='text-embedding-ada-002')