Skip to main content

Data analysis with evolutionary simulation

Project description

pikaia - Genetic AI

pikaia is the Python implementation of Genetic AI (evolutionary simulation for data analysis).

Installation

Use the package manager pip to install pikaia

pip install pikaia

Usage

We provide here the code for the "hello_model" example

import pikaia
import pikaia.alg
          

rawdata = np.zeros([3,3])
rawdata[:,:] = [[ 300, 10, 2],
                [ 600,  5, 2],
                [1500,  4, 1]]
# defines the variant fitness rules
gvfitnessrules = ["inv_percentage", "inv_percentage", "inv_percentage"]
# converts the raw data to a genetic population
data = pikaia.alg.Population(rawdata, gvfitnessrules)
# defines the used evolutionary strategies
strategy = ["GS Dominant", "OS Balanced"]
iterations = 1

# creating the genetic model
model = pikaia.alg.Model(data, strategy)

initialgenefitness = [1.0/3.0, 1.0/3.0, 1.0/3.0]
# returns the gene fitness values after 1 iteration
model.complete_run(initialgenefitness, iterations)

Examples

# provides the data for a small decision problem
example3x3 = pikaia.examples.assemble_example("3x3-DomBal+AltSal")

# provides the data for a real-world decision problem
example10x5 = pikaia.examples.assemble_example("10x5-DomBal+AltSal")

# use genetic ai to search a datafile using keywords and rank results
# for a more detailed example we refer to examples/geneticAI_run_search_example.py
search = pikaia.search.Search(data, orgs_labels, gens_labels)
fitnessOrganisms, fitnessGenes = search.search_request(query, top_k=5)

For details see examples/README.md.

Scientific Background

Please find the preprint of Genetic AI here

In Genetic AI, we convert a data problem to a model of genes and organisms. Afterwards, we run evolutionary simulations to obtain understanding of the input data.

Genetic AI is an AI that does not use training data to 'learn' but fully autonomously analyzes a problem. This is done by evolutionary strategies that cover certain 'behavior' and correlations of the input data.

License

MIT

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

pikaia-0.0.3.tar.gz (11.9 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

pikaia-0.0.3-py3-none-any.whl (13.7 kB view details)

Uploaded Python 3

File details

Details for the file pikaia-0.0.3.tar.gz.

File metadata

  • Download URL: pikaia-0.0.3.tar.gz
  • Upload date:
  • Size: 11.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.11.2

File hashes

Hashes for pikaia-0.0.3.tar.gz
Algorithm Hash digest
SHA256 1a98662243e5a8366cc7bdb0216ffcd398c3c03317b0a4e4d2c123a97dc561d1
MD5 a432ce1c9698772f1ba1235278012a68
BLAKE2b-256 75d5295766cdef4a402391b40dddd3c0d2f8ae06de6eab08b3d29995daaa2c8c

See more details on using hashes here.

File details

Details for the file pikaia-0.0.3-py3-none-any.whl.

File metadata

  • Download URL: pikaia-0.0.3-py3-none-any.whl
  • Upload date:
  • Size: 13.7 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.11.2

File hashes

Hashes for pikaia-0.0.3-py3-none-any.whl
Algorithm Hash digest
SHA256 4f95e90be6d2a0d2c90dfa6cd4dbb862d29689c09c32d4667d4a441bccf2f7d4
MD5 cb3e79d4bb0283d2b9bd45f75abfaa87
BLAKE2b-256 8c72e246f7f6bb738c70034be7d21c455c6ddbc629a1aade093e3781ffe42e3d

See more details on using hashes here.

Supported by

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