Supported deep learning frameworks
Supported deep learning frameworks included with IBM Watson® Machine Learning Accelerator.
The following frameworks are included with Watson Machine Learning Accelerator for use with IBM® Watson Machine Learning.
| Framework, library or asset | Version available in WML Accelerator 3.1.0 | Version available in WML Accelerator 3.0.0 |
|---|---|---|
| TensorFlow | 2.9.3 | 2.9.2 |
| PyTorch | 1.12.1 | 1.12.1 |
| Tensorboard | 2.9.1 | 2.9.1 |
| torchVision | 0.13.1 | 0.13.1 |
| OpenCV | 4.6.0 | 4.6.0 |
| scikit-learn | 1.1.1 | 1.1.1 |
| XGBoost | 1.6.2 | 1.6.2 |
| ONNX | 1.12.0 | 1.12.0 |
| PyArrow | 8.0.0 | 8.0.0 |
| Python | 3.10.6 | 3.10.4 |
- All frameworks support the supported versions of Python.
- PyTorch supports single node training, distributed training, and elastic distributed training.
- TensorFlow supports single node training, distributed training, and elastic distributed training (with TF.Keras).