imlresearch 1.0.0

A Python Toolkit for Reproducible Image-Based ML Experiments

ImageMLResearch

ImageMLResearch is a toolkit to help with image-based machine learning projects using Python. It includes functions for data handling, preprocessing, plotting, and more. These functions are combined into a single Researcher class to make experimentation easier and more efficient. Please note that this toolkit is specifically designed for image classification tasks and does not support regression problems.

For comprehensive usage instructions and API details, refer to the official documentation.

Installation

You can install ImageMLResearch using pip:

pip install imlresearch

upgrade to the latest version:

pip install --upgrade imlresearch

📦 Core Dependencies
This package supports Python 3.10–3.12. When installing, the following core libraries will also be installed:

numpy>=1.23.5,<2
tensorflow>=2.13,<2.18
optuna>=3.3,<4.8
opencv-python>=4.8,<4.14
scikit-learn>=1.4,<1.8
seaborn>=0.12.0,<=0.13.2

📦 Optional Dependency for AI Report Generation

openai==1.34.0

Install with:

pip install imlresearch[ai-report]

💡 Optional GPU Support for TensorFlow
If you have a compatible GPU and wish to enable GPU acceleration for TensorFlow, you can install the CUDA-enabled version with the following command:

pip install --cache-dir=/opt/tmp tensorflow[and-cuda]

🧪 Testing
The functionality of the code can be tested using the following command:

from imlresearch.api.tests import run_all_tests
run_all_tests()