-
State-of-the-art and not so state-of-the-art models trained with your own data with simple workflows.
-
Efficient data reading for (large) datasets in multiple formats and sources (CSV, Parquet, JSON, etc.).
-
Modular configuration and extensibility of models, datasets and training runs programmatically or via config files.
-
Use via
cli
or as plain Python (e.g., inside a Jupyter Notebook) -
Compatible with AllenNLP
For the installation we recommend setting up a fresh conda environment:
conda create -n biome python~=3.7.0 pip>=20.3.0
conda activate biome
Once the conda environment is activated, you can install the latest release via pip:
pip install -U biome-text
After installing biome.text, the best way to test your installation is by running the biome.text cli command:
biome --help
The best way to see how biome.text works is to go through our first tutorial.
Please refer to our documentation for more tutorials, detailed user guides and how you can contribute to biome.text.
The code in this project is licensed under Apache 2 license.