From 463c8fdbbce0c60eae2fc19397502794e4c4135c Mon Sep 17 00:00:00 2001
From: wassname <1103714+wassname@users.noreply.github.com>
Date: Fri, 6 Mar 2026 12:15:37 +0800
Subject: [PATCH] fix: apply Gemini review fixes (device kwarg, gradcheck
 requires_grad, torch prefix)

Review: Gemini 3.1 Pro approved. 3 fixes applied:
- pinn/SKILL.md: PchipFunction torch.tensor missing device=h.device (GPU crash)
- SKILL.md: gradcheck needs .requires_grad_(True) on doubled inputs
- SKILL.md: loss surface pseudocode now has torch. prefix + indexing='ij'
---
 SKILL.md      | 8 ++++----
 pinn/SKILL.md | 6 +++---
 2 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/SKILL.md b/SKILL.md
index 0ecabc8..3fc5576 100644
--- a/SKILL.md
+++ b/SKILL.md
@@ -93,7 +93,7 @@ logger.log("grad_norm", grad_norm)
 assert torch.isfinite(loss), f"Loss is {loss}"
 
 # Verify custom gradients (use float64! relative error plummets from 1e-2 to 1e-8)
-torch.autograd.gradcheck(my_custom_fn, inputs.double())
+torch.autograd.gradcheck(my_custom_fn, inputs.double().requires_grad_(True))
 ```
 
 Gradient clipping *masks* problems -- always log the pre-clip norm to see if it's constantly being triggered. [CS231n: "the ratio of the update magnitudes to the value magnitudes... should be somewhere around 1e-3."]
@@ -340,9 +340,9 @@ When a loss isn't behaving as expected, don't guess -- visualize the loss surfac
 ```py
 # ── 2D loss surface with gradient quiver ──────
 def analyze_component(loss_fn, x_range, y_range, n=80):
-    xs = linspace(*x_range, n)
-    ys = linspace(*y_range, n)
-    X, Y = meshgrid(xs, ys)
+    xs = torch.linspace(*x_range, n)
+    ys = torch.linspace(*y_range, n)
+    X, Y = torch.meshgrid(xs, ys, indexing='ij')
     x_flat = X.flatten().requires_grad_(True)
     y_flat = Y.flatten().requires_grad_(True)
 
diff --git a/pinn/SKILL.md b/pinn/SKILL.md
index 6167ff6..e0e3952 100644
--- a/pinn/SKILL.md
+++ b/pinn/SKILL.md
@@ -281,13 +281,13 @@ For heat exchangers with phase change, the T(h) mapping from REFPROP/GERG-2008 m
        def forward(ctx, h):
            T = pchip_interp(h.detach().numpy())  # scipy
            ctx.save_for_backward(h)
-           return torch.tensor(T, dtype=h.dtype)
+           return torch.tensor(T, dtype=h.dtype, device=h.device)
 
        @staticmethod
        def backward(ctx, grad_output):
            h, = ctx.saved_tensors
-           dTdh = pchip_interp.derivative()(h.detach().numpy())  # scipy
-           return grad_output * torch.tensor(dTdh, dtype=h.dtype)
+           dTdh = pchip_interp.derivative()(h.detach().cpu().numpy())  # scipy
+           return grad_output * torch.tensor(dTdh, dtype=h.dtype, device=h.device)
    ```
 5. Use float64 throughout. float32 loses precision near the phase boundary where dT/dh is small.