{"id":23295,"library":"autogluon-vision","title":"AutoGluon Vision","description":"AutoGluon Vision is part of the AutoGluon AutoML framework for computer vision tasks. It automates image classification, object detection, and other vision modalities. The latest release is v1.5.0 (Dec 2025). The package is under active development and follows the AutoGluon release cadence (~quarterly). Note: AutoGluon Vision is typically installed via the metapackage 'autogluon' or the standalone 'autogluon.vision' module; 'autogluon-vision' on PyPI is a legacy split package.","status":"active","version":"1.5.0","language":"python","source_language":"en","source_url":"https://github.com/autogluon/autogluon","tags":["autogluon","vision","image-classification","object-detection","automl","deep-learning","pytorch"],"install":[{"cmd":"pip install autogluon","lang":"bash","label":"Recommended: install full AutoGluon (includes vision)"},{"cmd":"pip install autogluon-vision","lang":"bash","label":"Legacy: install only the vision module (may be outdated)"}],"dependencies":[{"reason":"AutoGluon Vision uses PyTorch as the deep learning backend","package":"torch","optional":false},{"reason":"Required for image models and transforms","package":"torchvision","optional":false},{"reason":"Core library for training, prediction, and utilities","package":"autogluon.core","optional":false},{"reason":"Vision models are built on the multimodal framework since v1.0","package":"autogluon.multimodal","optional":false}],"imports":[{"note":"The module re-exports at the top level. Using the full path may fail if internal structure changes.","wrong":"from autogluon.vision.predictor import ImagePredictor","symbol":"ImagePredictor","correct":"from autogluon.vision import ImagePredictor"},{"note":"Same as above: prefer the top-level re-export.","wrong":"from autogluon.vision.detector import ObjectDetector","symbol":"ObjectDetector","correct":"from autogluon.vision import ObjectDetector"},{"note":"Since AutoGluon 0.8+, vision tasks are unified under multimodal. The old vision-specific predictor is deprecated.","wrong":"from autogluon.vision import MultiModalPredictor","symbol":"MultiModalPredictor","correct":"from autogluon.multimodal import MultiModalPredictor"}],"quickstart":{"code":"from autogluon.vision import ImagePredictor\n\npredictor = ImagePredictor()\npredictor.fit('https://autogluon.s3.amazonaws.com/datasets/shopee-iet.zip')\n\n# Predict on new images\nresults = predictor.predict('https://autogluon.s3.amazonaws.com/datasets/shopee-iet/test/')","lang":"python","description":"Train an image classification model on a sample dataset with minimal code. The fit method automatically downloads and preprocesses data."},"warnings":[{"fix":"Use Python >=3.9 (3.10–3.12 recommended). Check your environment with 'python --version'.","message":"AutoGluon 1.0 dropped support for Python 3.7. Python 3.8 support was dropped in v1.2. Use Python 3.9–3.12.","severity":"breaking","affected_versions":">=1.0.0"},{"fix":"Migrate to 'from autogluon.multimodal import MultiModalPredictor' and use the unified API for vision tasks.","message":"In AutoGluon 1.0, the 'autogluon.vision' module was unified under 'autogluon.multimodal'. Direct usage of old vision-specific APIs (e.g., 'ObjectDetector' without the multimodal wrapper) may break.","severity":"breaking","affected_versions":">=1.0.0"},{"fix":"Install 'pip install autogluon' to get the full suite (tabular, vision, text). This is the recommended approach since v1.0.","message":"The standalone 'autogluon-vision' package on PyPI may not be updated as frequently as the main 'autogluon' metapackage. Installing 'autogluon-vision' directly could result in an outdated version.","severity":"deprecated","affected_versions":"all"},{"fix":"Ensure your environment has a compatible GPU (NVIDIA with CUDA). Use 'import torch; print(torch.cuda.is_available())' to verify.","message":"AutoGluon Vision model training requires a GPU. CPU-only training may be extremely slow or fail due to insufficient memory.","severity":"gotcha","affected_versions":"all"},{"fix":"Always specify a local root path for data: predictor.fit('/path/to/local/data') or use a custom Dataset object.","message":"The 'fit()' method downloads datasets from S3 by default. Running in an offline or restricted network environment will fail unless you provide local data.","severity":"gotcha","affected_versions":"all"}],"env_vars":null,"last_verified":"2026-05-01T00:00:00.000Z","next_check":"2026-07-30T00:00:00.000Z","problems":[{"fix":"Run 'pip install autogluon' (not autogluon-vision). Then import via 'from autogluon.vision import ImagePredictor'.","cause":"Installed only 'autogluon-vision' but the import path changed; or installed an older version without the module.","error":"ModuleNotFoundError: No module named 'autogluon.vision'"},{"fix":"Specify a smaller model: predictor = ImagePredictor(); predictor.fit(..., hyperparameters={'model.mnames': ['resnet18']}).","cause":"The default model (e.g., ResNet) is too large for the GPU's memory. AutoGluon Vision automatically selects a model, but may exceed VRAM on smaller GPUs.","error":"RuntimeError: CUDA out of memory. Tried to allocate ..."},{"fix":"Check the documentation for the version you are using. Use 'predictor.fit(..., hyperparameters={'model': 'resnet18'})' for v1.x.","cause":"Incorrect hyperparameter name due to API changes between versions (e.g., 'arch' vs 'model').","error":"ValueError: Unknown hyperparameter '...'. Valid keys: ..."},{"fix":"Use: predictor.fit(..., hyperparameters={'optimization.max_epochs': 10})","cause":"The fit() API changed after v0.6. 'epochs' is not a direct argument; it's set via hyperparameters.","error":"TypeError: fit() got an unexpected keyword argument 'epochs'"}],"ecosystem":"pypi","meta_description":null,"install_score":null,"install_tag":null,"quickstart_score":null,"quickstart_tag":null}