MemoLib 0.1.0

A Python library for machine learning model training and inference

MemoLib

A Python library for machine learning model training and inference, supporting classification and object detection tasks.

Features

• Model Support: EfficientNet (B0-B7, V2), YOLO models
• Tasks: Classification and Object Detection
• Export: Support for ONNX and OpenVINO formats
• Training: Built-in training pipeline with callbacks
• Utilities: Dataset format conversion, custom loss functions

Installation

pip install MemoLib

Quick Start

from MemoLib import MemoModel, eModelTask, eClassifyModel

# Create a classification model
model = MemoModel(eModelTask.Classification, eClassifyModel.EfficientNetB0)

# Load weights and labels
model.LoadWeight("path/to/weights.pth")
model.LoadLabelName("path/to/labels.txt")

# Make predictions
result = model.Predict(image)
print(f"Prediction: {result.label}, Confidence: {result.confidence}")

Modules

• Model: Core model classes and interfaces
• DataSerializer: Configuration and serialization utilities
• DatasetFormatConvert: Dataset format conversion tools
• Loss: Custom loss functions (FocalLoss)

Requirements

• Python >= 3.8
• PyTorch >= 1.12.0
• OpenCV >= 4.5.0
• Other dependencies listed in pyproject.toml

License

MIT License