Orakl is an active learning library for TensorFlow.
System | 3.7 |
---|---|
Linux CPU |
Installation Requirements
- python >= 3.7
- tensorflow>=2.0.0
If you'd like to try our latest features, you can install the latest master directly from GitHub. For a basic install, run:
git clone https://github.com/nocotan/orakl.git
cd orakl
pip install -e .
To execute unit tests from a manual install, run:
python3 -m unittest tests/attr/core/strategies/test_expected_model_change.py
or recursive test execution via the command line is built-in:
python3 -m unittest discover tests
- Random Samplling (baseline)
- Expected Model Change Maximization [1]
- Margin Sampling [2]
Orakl is Apache-2.0 licensed, as found in the LICENSE file.
- [1] Cai, Wenbin, Ya Zhang, and Jun Zhou. "Maximizing expected model change for active learning in regression." 2013 IEEE 13th International Conference on Data Mining. IEEE, 2013.
- [2] Wang, Dan, and Yi Shang. "A new active labeling method for deep learning." 2014 International joint conference on neural networks (IJCNN). IEEE, 2014.