AutoGluon-Text
raw JSON → 0.6.2 verified Fri May 01 auth: no python deprecated
AutoGluon-Text is a submodule of AutoGluon for text classification and regression. It provides automated machine learning with multi-modal support (text, image, tabular). The latest PyPI version is 0.6.2, but the project has moved to a unified autogluon package (current version ~1.5.0). Release cadence is quarterly. Note: The autogluon.text package is deprecated; use autogluon>=1.0 and import from autogluon.multimodal instead.
pip install autogluon.text Common errors
error ModuleNotFoundError: No module named 'autogluon.text' ↓
cause autogluon.text not installed; also autogluon>=1.0 renamed the module.
fix
Install autogluon.text: pip install autogluon.text. For new code, use autogluon>=1.0: pip install autogluon; then import from autogluon.multimodal.
error ImportError: cannot import name 'TextPredictor' from 'autogluon' ↓
cause TextPredictor is not exported from the top-level autogluon package, it's in autogluon.text.
fix
Use: from autogluon.text import TextPredictor
Warnings
deprecated autogluon.text is deprecated in favor of autogluon.multimodal. New projects should use 'from autogluon.multimodal import MultiModalPredictor'. ↓
fix Switch to autogluon>=1.0 and use MultiModalPredictor from autogluon.multimodal.
gotcha pip install autogluon.text does not install the full autogluon suite. It only installs the text subpackage, which may have missing dependencies for image or tabular tasks. ↓
fix Install autogluon (full) for all features: 'pip install autogluon'.
breaking Python 3.10+ not supported in autogluon.text 0.6.2. Requires Python >=3.7,<3.10. ↓
fix Use Python 3.8 or 3.9, or upgrade to autogluon>=1.0 which supports Python 3.8-3.12.
deprecated The standalone autogluon.text package is no longer maintained. All future updates are in the unified autogluon package. ↓
fix Transition to 'pip install autogluon' and use MultiModalPredictor.
Imports
- TextPredictor wrong
from autogluon import TextPredictorcorrectfrom autogluon.text import TextPredictor
Quickstart
from autogluon.text import TextPredictor
# Download and train on a sample dataset
predictor = TextPredictor(label='label', eval_metric='acc')
predictor.fit('https://autogluon.s3.amazonaws.com/datasets/Inc/train.csv')
# Predict on test data
predictions = predictor.predict('https://autogluon.s3.amazonaws.com/datasets/Inc/test.csv')
print(predictions.head())