Skip to main content

SensoPy

SPEC 0 Tests codecov

Python library for the design and analysis of sensory discrimination tests. I have plans to make this a full-fledged sensory data analysis library by including all common standard techniques (Preference Mapping, Penalty Analysis, DOE, etc.)

Installation

pip install SensoPy

Usage

Difference test

One-tailed test of H0: p_d = 0 (no difference) against H1: p_d > 0 (a difference exists). 19 correct responses out of 30 panelists using the Triangle method:

from sensopy import DiscriminationTest
from sensopy.discrimination import TRIANGLE

test = DiscriminationTest(TRIANGLE)
result = test.difference(19, 30)

print(f"d'      = {result.d_prime.estimate:.4f}  [{result.d_prime.lower:.4f}, {result.d_prime.upper:.4f}]")
print(f"p_d     = {result.pd.estimate:.4f}  [{result.pd.lower:.4f}, {result.pd.upper:.4f}]")
print(f"p-value = {result.p_value:.4f}")
d'      = 2.1462  [1.1262, 3.1336]
p_d     = 0.4500  [0.1578, 0.7011]
p-value = 0.0007
  • d' = 2.15 (95% CI: [1.13, 3.13]): estimated Thurstonian discriminal distance, a method-independent measure of the sensory difference between the two products.
  • p_d = 0.45 (95% CI: [0.16, 0.70]): an estimated 45% of the panel can reliably detect the difference.
  • p-value = 0.0007: strong evidence against H0; the products are perceptibly different at any conventional significance level.

Equivalence test

One-tailed test of H0: p_d >= 0.30 against H1: p_d < 0.30 (the products are sensorially similar). pd0 = 0.30 sets the equivalence threshold: the maximum proportion of discriminators we are willing to tolerate and still call the products equivalent. 16 correct out of 30 panelists using the 2-AFC method:

from sensopy import DiscriminationTest
from sensopy.discrimination import TWO_AFC

test = DiscriminationTest(TWO_AFC)
result = test.equivalence(16, 30, pd0=0.30)

print(f"d'      = {result.d_prime.estimate:.4f}  [{result.d_prime.lower:.4f}, {result.d_prime.upper:.4f}]")
print(f"p_d     = {result.pd.estimate:.4f}  [{result.pd.lower:.4f}, {result.pd.upper:.4f}]")
print(f"p-value = {result.p_value:.4f}")
d'      = 0.1183  [0.0000, 0.8099]
p_d     = 0.0667  [0.0000, 0.4332]
p-value = 0.1263
  • d' = 0.12 (95% CI: [0.00, 0.81]): small estimated sensory distance, consistent with near-identical products.
  • p_d = 0.07 (95% CI: [0.00, 0.43]): only ~7% of the panel estimated to reliably discriminate the products.
  • p-value = 0.13: insufficient evidence to conclude equivalence at α = 0.05 with pd0 = 0.30 (i.e., we cannot rule out that ≥ 30% of consumers can detect the difference); a larger panel would be needed to meet that threshold.

Roadmap

See ROADMAP.md.

Citing

If you use SensoPy in your work, please cite it using CITATION.cff; GitHub's "Cite this repository" button exports it as APA or BibTeX.

Credits

  • Edgar Ramírez Mondragón

References

  1. Bi, J. 2015. Sensory Discrimination Tests and Measurements. Sensometrics in Sensory Evaluation. Wiley Blackwell, Richmond, Virginia, USA.

Metadata

Release files for sensopy 0.0.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for sensopy 0.0.2
File Size Uploaded Provenance Actions
sensopy-0.0.2.tar.gz 59.8 kB With provenance Details for sensopy-0.0.2.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for sensopy 0.0.2
File Interpreter ABI Platform Actions
sensopy-0.0.2-py3-none-any.whl Python 3 none any Details for sensopy-0.0.2-py3-none-any.whl

Total release size: 72.6 kB

Release files / sensopy-0.0.2.tar.gz

Download URL sensopy-0.0.2.tar.gz
Size 59.8 kB
Tags Source
SHA-256 checksum
How to use checksums
bcd6882694e644af918696a83fd5ff8c138ae0d27e64108570426ee38391f645
BLAKE2b-256 checksum
How to use checksums
6de6df5f192eada8cff7a240b5bd50261b34d11e5e1b162434af14065332f18a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 10, 2026.

Transparency log

Release files / sensopy-0.0.2-py3-none-any.whl

Download URL sensopy-0.0.2-py3-none-any.whl
Size 12.8 kB
Tags Python 3
SHA-256 checksum
How to use checksums
b60b06f1598592f17eae0effdc2d7ddc53336e610b71522b1a26f5e7c5d14247
BLAKE2b-256 checksum
How to use checksums
4e8ae06a67333f60a3b2410a1ece7f61dee8ba2b0268475c233f6c642027bdf0
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 10, 2026.

Transparency log

Security for sensopy 0.0.2

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for sensopy-0.0.2.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 10, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page