README.md in torch-rb-0.5.2 vs README.md in torch-rb-0.5.3
- old
+ new
@@ -6,11 +6,11 @@
- [TorchVision](https://github.com/ankane/torchvision) for computer vision tasks
- [TorchText](https://github.com/ankane/torchtext) for text and NLP tasks
- [TorchAudio](https://github.com/ankane/torchaudio) for audio tasks
-[](https://travis-ci.org/ankane/torch.rb)
+[](https://github.com/ankane/torch.rb/actions)
## Installation
First, [install LibTorch](#libtorch-installation). For Homebrew, use:
@@ -317,10 +317,17 @@
net = MyNet.new
net.load_state_dict(Torch.load("net.pth"))
net.eval
```
+When saving a model in Python to load in Ruby, convert parameters to tensors (due to outstanding bugs in LibTorch)
+
+```python
+state_dict = {k: v.data if isinstance(v, torch.nn.Parameter) else v for k, v in state_dict.items()}
+torch.save(state_dict, "net.pth")
+```
+
### Tensor Creation
Here’s a list of functions to create tensors (descriptions from the [C++ docs](https://pytorch.org/cppdocs/notes/tensor_creation.html)):
- `arange` returns a tensor with a sequence of integers
@@ -414,10 +421,10 @@
Here’s the list of compatible versions.
Torch.rb | LibTorch
--- | ---
-0.5.0+ | 1.7.0
+0.5.0+ | 1.7.0-1.7.1
0.3.0+ | 1.6.0
0.2.0-0.2.7 | 1.5.0-1.5.1
0.1.8 | 1.4.0
0.1.0-0.1.7 | 1.3.1