torch-rb 0.17.1 → 0.19.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f04aeddd9c66d668f78ed38d831ec66907743fd5be2cd076e9e547d00f1f2aa3
-  data.tar.gz: e76ca5199c4801c0eaa760ed3e41ebb6e521774bba084809656b1ed9e4b96483
+  metadata.gz: 03164cc479d8f8a32f0669d597e8fe5310d91955e6954cfdc0fffdc8983c5768
+  data.tar.gz: 87fc733016b6f4489b38a419a3879cacbdb1e190cfaa5c02397aceb57c012d16
 SHA512:
-  metadata.gz: 683246aa91d19daa624a9d3a89fa5e4898f19fbb45a72ff2bcac0ee3025859f0f8ad26ac153e58150aedf10547cf670e2fe131cffda246546ded1f1344637040
-  data.tar.gz: bd0b955b997f09a274d8b59a21eafa8bf45972b6a20883b9d457a29417aa2c6d413b324abadfb6772c5bdc3aa0e633d3e25d6f23dd986fa603c55c69ae4a5a85
+  metadata.gz: 6ba0480138a10ba43dff625dc1bcf99e2287f238dc4607ea6813e82914f1e335133f55408fa59579343b161c54850316f744c594cf6687b7f2de64a0d71746d1
+  data.tar.gz: 859015641dd14bf919a7982c6673acb296f858518552b8c924fc7e59b9c1b2a9491aa598c01b019b392c8c2bba7b9f65ff0923f838e6cbde7ddaede9c4b69191

data/CHANGELOG.md

@@ -1,3 +1,14 @@
+## 0.19.0 (2025-01-29)
+
+- Updated LibTorch to 2.6.0
+- Improved `inspect` for `Device`
+- Fixed equality for `Device`
+- Fixed `index` method for `Device` when no index
+
+## 0.18.0 (2024-10-22)
+
+- Updated LibTorch to 2.5.0
+
 ## 0.17.1 (2024-08-19)
 
 - Added `persistent` option to `register_buffer` method

data/README.md

@@ -14,13 +14,22 @@ Check out:
 
 ## Installation
 
-First, [install LibTorch](#libtorch-installation). With Homebrew, it’s part of the PyTorch package:
+First, [download LibTorch](https://pytorch.org/get-started/locally/). For Mac arm64, use:
 
 ```sh
-brew install pytorch
+curl -L https://download.pytorch.org/libtorch/cpu/libtorch-macos-arm64-2.6.0.zip > libtorch.zip
+unzip -q libtorch.zip
 ```
 
-Add this line to your application’s Gemfile:
+For Linux x86-64, use the `cxx11 ABI` version. For other platforms, build LibTorch from source.
+
+Then run:
+
+```sh
+bundle config build.torch-rb --with-torch-dir=/path/to/libtorch
+```
+
+And add this line to your application’s Gemfile:
 
 ```ruby
 gem "torch-rb"
@@ -398,18 +407,14 @@ Here’s a list of functions to create tensors (descriptions from the [C++ docs]
   Torch.zeros(3) # tensor([0, 0, 0])
   ```
 
-## LibTorch Installation
-
-[Download LibTorch](https://pytorch.org/) (for Linux, use the `cxx11 ABI` version). Then run:
-
-```sh
-bundle config build.torch-rb --with-torch-dir=/path/to/libtorch
-```
+## LibTorch Compatibility
 
 Here’s the list of compatible versions.
 
 Torch.rb | LibTorch
 --- | ---
+0.19.x | 2.6.x
+0.18.x | 2.5.x
 0.17.x | 2.4.x
 0.16.x | 2.3.x
 0.15.x | 2.2.x
@@ -417,14 +422,6 @@ Torch.rb | LibTorch
 0.13.x | 2.0.x
 0.12.x | 1.13.x
 
-### Homebrew
-
-You can also use Homebrew.
-
-```sh
-brew install pytorch
-```
-
 ## Performance
 
 Deep learning is significantly faster on a GPU.

data/codegen/generate_functions.rb

@@ -53,7 +53,9 @@ def skip_functions(functions)
     f.base_name == "sym_size" ||
     f.base_name == "sym_numel" ||
     f.base_name == "sym_storage_offset" ||
-    f.base_name == "sym_stride"
+    f.base_name == "sym_stride" ||
+    # TODO fix LibTorch 2.6 changes
+    f.base_name == "rrelu_with_noise"
   end
 end