Mention AMD

Co-authored-by: Steven Palma <[email protected]>
Signed-off-by: Vladislav Sovrasov <[email protected]>

Author: sovrasov

docs/source/torch_accelerators.mdx

@@ -4,7 +4,7 @@ LeRobot supports multiple hardware acceleration options for both training and in
 
 These options include:
 - **CPU**: CPU executes all computations, no dedicated accelerator is used
-- **CUDA**: acceleration with NVIDIA GPUs
+- **CUDA**: acceleration with NVIDIA & AMD GPUs
 - **MPS**: acceleration with Apple Silicon GPUs
 - **XPU**: acceleration with Intel integrated and discrete GPUs