getitune 0.1.2

getitune: Train, Evaluate, Optimize, Deploy Computer Vision Models

A low-code transfer learning framework for training, evaluating, optimizing, and deploying computer vision models

---

Key Features • Supported Tasks & Models • Installation • Quick Start • Docs • License

---

Geti Library - getitune

The Geti™ library (getitune) is a low-code transfer learning framework for Computer Vision. Its API and CLI let you train, evaluate, optimize, and deploy models quickly, even without deep expertise in deep learning. It supports diverse combinations of model architectures, learning methods, and task types built on PyTorch and the OpenVINO™ toolkit.

Each supported task ships with curated "recipes": YAML files that bundle the model, data pipeline, and training configuration into a single one-stop entry point. Recipes are validated on standard datasets so you get a strong baseline out of the box.

Key Features

• Multi-task support: classification, object detection, rotated detection, instance segmentation, semantic segmentation, and keypoint detection, see the full model list below.
• Intel Hardware acceleration: Native Intel GPU (XPU) support.
• Tiling for large images across detection and segmentation tasks.
• Multiple backends: train with PyTorch Lightning, export and run inference with ONNX and OpenVINO™.
• Datumaro data frontend with automatic format detection (COCO, YOLO, VOC, native).
• Distributed training across multiple GPUs.
• Mixed-precision training to reduce memory and increase batch size.
• Class-incremental learning to extend an existing model with new classes.
• Deployment to OpenVINO™ IR and ONNX formats, with inference via OpenVINO™ ModelAPI.

---

Supported Tasks & Models

All recipes live under src/getitune/recipe/<task>/. Pass any of these YAMLs directly to the API as model=.... Recipes whose name ends in _tile enable the tiling pipeline for large images.

Task	Recipe directory	Example recipes
Classification (multi-class / multi-label / hierarchical)	multi_class_cls, multi_label_cls, h_label_cls	dino_v2, vit_tiny, efficientnet_b0, efficientnet_b3, efficientnet_v2, mobilenet_v3_large
Object detection	detection	atss_mobilenetv2, ssd_mobilenetv2, yolox_{tiny,s,l,x}, rtdetr_50, dfine_x, deim_dfine_{l,m,x}, deimv2_{s,m,l}, rfdetr_{small,medium,large}
Rotated detection	rotated_detection	maskrcnn_r50, maskrcnn_efficientnetb2b (with _tile variants)
Instance segmentation	instance_segmentation	maskrcnn_{r50,swint,efficientnetb2b}, rtmdet_inst_tiny, rfdetr_seg_{small,medium,large,xlarge}
Semantic segmentation	semantic_segmentation	dino_v2, litehrnet_{s,18,x}, segnext_{t,s,b} (with _tile variants)
Keypoint detection	keypoint_detection	rtmpose_tiny

Each task directory also ships an openvino_model.yaml recipe for running and optimizing pre-exported OpenVINO IR models via OVEngine.

---

Installation

Requirements: Python 3.11–3.14, PyTorch 2.10, OpenVINO™ 2026.1, NumPy ≥ 2.0.

getitune is published on PyPI.

Quick Install

# CPU-only (default, works on all platforms)
uv pip install getitune

# Or with pip
pip install getitune

Advanced Installation: Specify Hardware Backend

getitune ships three mutually exclusive extras that select the right PyTorch wheel for your hardware:

Extra	PyTorch wheel	Use when
[xpu]	torch==2.10.0+xpu + triton-xpu	Intel discrete or integrated GPUs.
[cuda]	torch==2.10.0+cu128	NVIDIA GPUs with CUDA 12.8 drivers.
[cpu]	torch==2.10.0+cpu (Linux/Windows) or default	No GPU, or running on Apple silicon.

Since PyTorch distributes GPU wheels separately, you must include the PyTorch index:

# Intel GPU (XPU)
uv pip install "getitune[xpu]" \
  --extra-index-url https://download.pytorch.org/whl/xpu

# NVIDIA GPU (CUDA 12.8)
uv pip install "getitune[cuda]" \
  --extra-index-url https://download.pytorch.org/whl/cu128

# CPU-only (no extra index needed)
uv pip install "getitune[cpu]"

macOS note: PyTorch's +cpu wheel is only published for Linux and Windows. The [cpu] extra resolves this automatically and installs the default torch==2.10.0 wheel on macOS.

Install from Source

git clone https://github.com/open-edge-platform/training_extensions.git
cd training_extensions/library

# Recommended: use uv to honor the lockfile
uv sync --extra xpu          # or --extra cpu / --extra cuda

---

Quick Start

Training and Exporting

Getitune supports an API-based training approach. To run training with data config and lightning backend:

from getitune.backend.lightning.engine import LightningEngine

# Initialize and train using the bundled test dataset
engine = LightningEngine(
    model="src/getitune/recipe/classification/multi_class_cls/efficientnet_b0.yaml", # path to recipe YAML / model class instance / model name (e.g., "efficientnet_b0")
    data_root="tests/assets/classification_cifar10", # path to dataset (any supported format, e.g., COCO, VOC, YOLO, Datumaro)
)

It is also possible to create an engine with model name and task type. Getitune will look for a matching recipe in the bundled recipes and use it to set up the training configuration:

engine = LightningEngine(
    model="efficientnet_b0", # model name (e.g., "efficientnet_b0")
    task="MULTI_CLASS_CLS", # task type (e.g., "MULTI_CLASS_CLS")
    data_root="tests/assets/classification_cifar10", # path to dataset (any supported format, e.g., COCO, VOC, YOLO, Datumaro)
)

Run training, test, predict and export the model to OpenVINO IR format:

if __name__ == "__main__": # to avoid issues with multiprocessing
  engine.train()
  metrics = engine.test() # validate on test set
  predictions = engine.predict() # predict on test set
  exported_ov_path = engine.export()  # writes OpenVINO IR

Geti Library also supports export to ONNX format, which can be enabled by passing export format when calling export(). To specify export precision, use the export_precision argument. By default, models are exported in FP32 precision.

from getitune.types.export import ExportFormat, ExportPrecision

exported_onnx_path = engine.export(export_format=ExportFormat.ONNX, export_precision=ExportPrecision.FP16)  # writes ONNX model in FP16 precision

Inference and Optimization

Getitune provides inference via PyTorch and OpenVINO / ONNX backends:

from getitune.backend.openvino.engine import OVEngine

# PyTorch inference
engine = OVEngine(data="/path/to/dataset", model=exported_ov_path) # can also pass onnx model here
metrics = engine.test() # test OpenVINO model accuracy
predictions = engine.predict() # predict on test set

OpenVINO backend provides optimization capabilities like post-training quantization via NNCF. Post-training quantization is supported only for OpenVINO IR models.

ov_engine = OVEngine(data="/path/to/dataset", model="path/to/exported_model.xml")
ov_engine.optimize()
ov_engine.test() # test optimized model accuracy
predictions = ov_engine.predict() # predict with optimized model

---

Dataset Support

When you pass a path to data=, getitune uses Datumaro to auto-detect the dataset format. Supported formats:

Format	Detection method
COCO	annotations/ directory with COCO JSON files
YOLO	data.yaml file (Ultralytics layout)
Pascal VOC	JPEGImages/, Annotations/, ImageSets/ directories
Datumaro (native)	metadata.json + data.parquet at root

Zip archives are also accepted, Datumaro extracts them on import.

# Works the same regardless of format, just point to the dataset root
engine = create_engine(
    data="/path/to/coco_or_yolo_or_voc_dataset",
    model=model,  # a model class instance, or an exported OpenVINO/ONNX model path
)
engine.train()

---

Advanced Usage

Override training hyperparameters

You can override engine-level training parameters directly:

engine.train(
    max_epochs=50,
    seed=42,
    deterministic=True,
    precision="16-mixed",          # mixed-precision training
    gradient_clip_val=1.0,
    check_val_every_n_epoch=5,
)

For model-level hyperparameters like learning rate and optimizer, you can pass them when instantiating a model class directly:

from torch.optim import AdamW
from getitune.models import EfficientNet

model = EfficientNet(
    label_info=datamodule.label_info, # or simply num_classes, e.g., int value
    model_name="efficientnet_b0",
    optimizer=lambda params: AdamW(params, lr=0.001, weight_decay=0.01),
)

engine = LightningEngine(data=datamodule, model=model)
engine.train(max_epochs=100)

Alternatively, you can set these in a custom recipe YAML (copy and modify an existing one):

# my_recipe.yaml — custom learning rate and optimizer
task: MULTI_CLASS_CLS
model:
  class_path: getitune.backend.lightning.models.classification.multiclass_models.efficientnet.EfficientNetMulticlassCls
  init_args:
    label_info: 1000
    model_name: efficientnet_b0

    optimizer:
      class_path: torch.optim.AdamW
      init_args:
        lr: 0.001
        weight_decay: 0.01

    scheduler:
      class_path: getitune.backend.lightning.schedulers.LinearWarmupSchedulerCallable
      init_args:
        num_warmup_steps: 5
        main_scheduler_callable:
          class_path: lightning.pytorch.cli.ReduceLROnPlateau
          init_args:
            mode: max
            factor: 0.5
            patience: 3
            monitor: val/accuracy

data: src/getitune/recipe/_base_/data/classification.yaml

overrides:
  max_epochs: 100

Override augmentations and datamodule

For augmentations, override the data config. Augmentations run on CPU (augmentations_cpu) and GPU (augmentations_gpu) separately:

# my_data.yaml — stronger augmentations
task: MULTI_CLASS_CLS
input_size: [224, 224]
train_subset:
  subset_name: train
  batch_size: 32
  num_workers: 8
  augmentations_cpu:
    - class_path: torchvision.transforms.v2.RandomResizedCrop
      init_args:
        size: [224, 224]
        scale: [0.08, 1.0]
  augmentations_gpu:
    - class_path: kornia.augmentation.RandomHorizontalFlip
      init_args:
        p: 0.5
    - class_path: kornia.augmentation.ColorJiggle
      init_args:
        brightness: 0.4
        contrast: 0.4
        saturation: 0.4
        hue: 0.1
        p: 0.8
    - class_path: kornia.augmentation.RandomGaussianBlur
      init_args:
        kernel_size: [3, 3]
        sigma: [0.1, 2.0]
        p: 0.3
    - class_path: kornia.augmentation.Normalize
      init_args:
        mean: [0.485, 0.456, 0.406]
        std: [0.229, 0.224, 0.225]

Augmentations can be overridden using API as well when creating a DataModule instance and passing it to the engine:

from getitune.data.datamodule import DataModule
from getitune.config.data import SubsetConfig
import kornia.augmentation as K
from torchvision.transforms import v2

datamodule = DataModule(
    task = "MULTI_CLASS_CLS",
    data_root = "library/tests/assets/classification_cifar10",
    train_subset = SubsetConfig(
        augmentations_cpu=[
            v2.Resize((256, 256)),
        ],
        augmentations_gpu=[
            K.RandomErasing(p=0.5),
            K.ColorJitter(brightness=0.2, contrast=0.2, saturation=0.2, hue=0.1),
        ],
        batch_size=32,
        num_workers=8
    ),
    val_subset = SubsetConfig(
        subset_name="val",
        augmentations_cpu=[
            v2.Resize((256, 256)),
        ],
        batch_size=32,
        num_workers=8
    ),
    test_subset = SubsetConfig(
        subset_name="test",
        augmentations_cpu=[
            v2.Resize((256, 256)),
        ],
        batch_size=32,
        num_workers=8
    ),
)

model = EfficientNet(label_info=datamodule.label_info, model_name="efficientnet_b0")
engine = LightningEngine(data=datamodule, model=model)
engine.train()

Available model classes:

• Detection: ATSS, SSD, YOLOX, RTDETR, DFine, DEIMDFine, DEIMV2, RFDETR
• Instance segmentation: MaskRCNN, MaskRCNNTV, RTMDetInst, RFDETRSeg
• Semantic segmentation: DinoV2Seg, LiteHRNet, SegNext
• Classification: EfficientNet, MobileNetV3, VisionTransformer, TimmModel, TVModel
• Keypoint: RTMPose

---

License

The Geti™ Library (getitune) is licensed under Apache License Version 2.0. By contributing to the project, you agree to the license and copyright terms therein and release your contribution under these terms.

---