Remove duplicate sentences in description of torch.linalg.eig (#108230)
This removes nearly identical sentences in the description of `torch.linalg.eig` describing the checks in the backward pass by @lezcano
Pull Request resolved: https://github.com/pytorch/pytorch/pull/108230
Approved by: https://github.com/lezcano
diff --git a/torch/linalg/__init__.py b/torch/linalg/__init__.py
index 5192431..eddaf09 100644
--- a/torch/linalg/__init__.py
+++ b/torch/linalg/__init__.py
@@ -477,9 +477,6 @@
of the matrix. For this reason, the loss function shall not depend on the phase of the
eigenvectors, as this quantity is not well-defined.
This is checked when computing the gradients of this function. As such,
- when inputs are on a CUDA device, this function synchronizes that device with the CPU
- when computing the gradients.
- This is checked when computing the gradients of this function. As such,
when inputs are on a CUDA device, the computation of the gradients
of this function synchronizes that device with the CPU.
