BlackHC/tfpyth
A library enabling differentiable interoperability between TensorFlow and PyTorch.
Not currently ranked — collecting fresh signals.
star history
TfPyTh provides functions to wrap TensorFlow graphs as callable PyTorch functions and vice versa. It allows combining codebases from both frameworks into single trainable models with full gradient support. Tensors are copied through CPU memory since the frameworks use incompatible GPU array interfaces.
Frequently asked
- What is BlackHC/tfpyth?
- A library enabling differentiable interoperability between TensorFlow and PyTorch.
- Is tfpyth open source?
- Yes — BlackHC/tfpyth is open source, released under the MIT license.
- What language is tfpyth written in?
- BlackHC/tfpyth is primarily written in Python.
- How popular is tfpyth?
- BlackHC/tfpyth has 646 stars on GitHub.
- Where can I find tfpyth?
- BlackHC/tfpyth is on GitHub at https://github.com/BlackHC/tfpyth.