Skip to main content

Explainable AI, Model Monitoring, and Outlier Detection Tools for Computer Vision Systems

Project description

Obz AI 🔍: Explainable AI, Model Monitoring, and Outlier Detection for Computer Vision Systems

Obz AI is a powerful Python package designed to bring explainability, continuous monitoring, and advanced outlier detection to AI-powered computer vision systems. With support for the latest deep learning architectures -- including vision transformers (ViT) and convolutional neural networks (CNN) -- Obz AI enables ML practitioners and data scientists to ensure transparency, reliability, and trustworthiness in their vision models.

Obz AI offers seamless integration, allowing you to use its robust features as a standalone tool or connect effortlessly to the dashboard for real-time model monitoring, data visualization, and configuration. Track your machine learning models, visualize image data, generate XAI (Explainable AI) heatmaps, and perform anomaly detection.

For more details, demo, and full documentation, visit Obz.AI.

Key Features

  • Data Inspector Module: Automatically extracts features and detects outliers, data drifts, or other anomalies in image datasets using advanced statistical and machine learning methods — improving data quality and model robustness.
  • XAI Module: Generates state-of-the-art explainability heatmaps for computer vision models, including Saliency Maps, Attention Maps, CDAM, and more. Provides quantitative evaluation tools such as fidelity and compactness for model interpretability and explainability.
  • Obz Client: Effortlessly log and monitor your models, inputs, outputs, and XAI explanations to the Obz AI server (cloud or on-prem) for comprehensive oversight and user-friendly visualization.

Obz AI is the all-in-one solution for anyone building, deploying, or monitoring computer vision models -- empowering explainable, reliable, and secure AI applications.

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

obzai-0.2.0.tar.gz (60.6 kB view details)

Uploaded Source

Built Distribution

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

obzai-0.2.0-py3-none-any.whl (97.2 kB view details)

Uploaded Python 3

File details

Details for the file obzai-0.2.0.tar.gz.

File metadata

  • Download URL: obzai-0.2.0.tar.gz
  • Upload date:
  • Size: 60.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for obzai-0.2.0.tar.gz
Algorithm Hash digest
SHA256 562c537d4fff0931be49ffb6504e081b3371e74608cba1c408a27e49cc3555f2
MD5 25615a9d593cd55d25addbe9b3de9ec9
BLAKE2b-256 a3b189c38128b266bc5a0b83713995fe4ab3e84e5eb15b783afa4626ebb8a01f

See more details on using hashes here.

Provenance

The following attestation bundles were made for obzai-0.2.0.tar.gz:

Publisher: release.yaml on alethia-xai/obzai

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file obzai-0.2.0-py3-none-any.whl.

File metadata

  • Download URL: obzai-0.2.0-py3-none-any.whl
  • Upload date:
  • Size: 97.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for obzai-0.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 b5d296800816b986b607ddf37fdbfa3f91d16b445f93ff4516c24ee8875d5f60
MD5 a49b1c31d167ca17ed05d8e2533cd29f
BLAKE2b-256 97d46882bb91b302531611c1fa9a4832606a2e89ad977905d25c9abffa414a90

See more details on using hashes here.

Provenance

The following attestation bundles were made for obzai-0.2.0-py3-none-any.whl:

Publisher: release.yaml on alethia-xai/obzai

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

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