Skip to main content

Kiri Natural Language Engine

Project description

Kiri Natural Language Engine

Kiri NLE is a high level Python library that makes using state-of-the-art language models easy, accessible and scalable.

With Kiri, no experience in AI is needed to solve valuable real world problems using:

  • Semantic search in 50+ languages (for ecommerce, documentation, etc.)
  • Conversational question answering in English (for FAQ chatbots, text analysis, etc.)
  • Zero-shot classification in 100+ languages (for email sorting, intent detection, etc.)
  • Summarisation in English (TLDRs for long documents)
  • Emotion detection in English (for customer satisfaction, text analysis, etc.)

Run everything locally or take your code to production using our optimised inference API, where you only pay for usage.

Getting started Installation, few minute introduction
💡 Examples Sample problems solved using Kiri
📙 Docs In-depth documentation for advanced usage

Getting started

Installation

Install Kiri via PyPi:

pip install kiri

Basic usage

from kiri import Kiri, Document

# Unprocessed documents
documents = [
    Document("Look at examples to see awesome use cases!"),
    Document("Check out the docs to see what's possible!")
]

# Use our inference API
kiri = Kiri(api_key="abc")
# Or run locally
kiri = Kiri(local=True)

# Process documents
kiri.upload(documents)

# Start building!
search_results = kiri.search("What are some cool apps that have been built?")

print(search_results.to_json())

# Prints
{
   "max_score": 0.3804888461635889,
   "total_results": 2,
   "results": [
      {
         "document": {
            "id":"LzhtWcpV2eoMk8GJwaw7na",
            "content":"Look at examples to see awesome use cases!"
         },
         "score": 0.3804888461635889,
         "preview":" Look at examples to see awesome use cases!"
      },
      {
         "document": {
            "id":"bcLb8xUK585Zm6rZrwj89A",
            "content":"Check out the docs to see what's possible!"
         },
         "score": 0.1742559312454076,
         "preview":" Check out the docs to see what's possible!"
      }
   ]
}

Why Kiri?

  1. No experience needed

    • Entrance to practical AI should be simple
    • Get state-of-the-art performance in your task without being an expert
  2. There is an overwhelming amount of models

    • We implement the best ones for various use cases
    • A few general models can accomplish more with less optimisation
  3. Deploying models cost effectively is hard work

    • If our models suit your use case, no deployment is needed
    • Our API scales, is always available, and you only pay for usage

Examples

Take a look at the examples folder.

Documentation

Check out our docs.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

kiri-0.4.1.tar.gz (19.1 kB view details)

Uploaded Source

Built Distribution

kiri-0.4.1-py3-none-any.whl (49.2 kB view details)

Uploaded Python 3

File details

Details for the file kiri-0.4.1.tar.gz.

File metadata

  • Download URL: kiri-0.4.1.tar.gz
  • Upload date:
  • Size: 19.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.2.0 pkginfo/1.6.1 requests/2.22.0 setuptools/45.2.0 requests-toolbelt/0.9.1 tqdm/4.55.0 CPython/3.8.5

File hashes

Hashes for kiri-0.4.1.tar.gz
Algorithm Hash digest
SHA256 494ab38c8f94d68db472f370623d622ad7ef12a06c10abad9fdc06155510c2e8
MD5 f52144cdba6723e865139d9be8aa9e3b
BLAKE2b-256 2670a70d50b0b7caea0157510f0adcf2a01f82b623fbf6cece3ca72f430dba26

See more details on using hashes here.

File details

Details for the file kiri-0.4.1-py3-none-any.whl.

File metadata

  • Download URL: kiri-0.4.1-py3-none-any.whl
  • Upload date:
  • Size: 49.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.2.0 pkginfo/1.6.1 requests/2.22.0 setuptools/45.2.0 requests-toolbelt/0.9.1 tqdm/4.55.0 CPython/3.8.5

File hashes

Hashes for kiri-0.4.1-py3-none-any.whl
Algorithm Hash digest
SHA256 e10437d3483fa1347b26c6de4f6350723463eec47019b4f0b11e68a1e205c274
MD5 933c32e7a2758fbfa251fe087ce91f3a
BLAKE2b-256 2e3954350106735fdb330f99432957098a6f4a3993d4106ba0819d40cbdab307

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page