To be clear, the goal is not to reinsert in the codebase, but make it so it can be made to work together when users install an extra package on top of PyTensor. We are still not interested in supporting GPU directly due to little developer capacity
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Theano deprecation at 1.0 | 16 | 4167 | October 7, 2017 | |
| Setup environment for using GPU with Theano | 1 | 3409 | July 24, 2018 | |
| Tensorflow backend | 2 | 992 | September 19, 2017 | |
| Using pycuda in theano operation | 2 | 1241 | June 28, 2019 | |
| Leveraging the GPU in PyMC3 | 5 | 8052 | August 12, 2017 |