Silent validation

examples/rtc/eval_dataset.py

@@ -18,6 +18,7 @@ Usage:
         --rtc.execution_horizon=8 \
         --device=mps \
         --rtc.max_guidance_weight=10.0 \
+        --rtc.prefix_attention_schedule=ONES \
         --seed=10
 
     # Basic usage with pi0.5 policy
@@ -512,9 +513,9 @@ class RTCEvaluator:
         self.plot_tracked_data(rtc_tracked_steps, no_rtc_tracked_steps, prev_chunk_left_over, num_steps)
 
         # Validate RTC behavior
-        logging.info("=" * 80)
+        # logging.info("=" * 80)
-        logging.info("Validating RTC behavior...")
+        # logging.info("Validating RTC behavior...")
-        self.validate_rtc_behavior(rtc_actions, no_rtc_actions, prev_chunk_left_over)
+        # self.validate_rtc_behavior(rtc_actions, no_rtc_actions, prev_chunk_left_over)
 
         # Plot final actions comparison
         logging.info("=" * 80)
@@ -800,6 +801,9 @@ class RTCEvaluator:
             num_steps,
         )
 
+        # Plot no-RTC x_t data on right chart as orange dashed line for comparison
+        self._plot_no_rtc_xt_reference(no_rtc_tracked_steps, axs_xt[:, 1], num_steps)
+
         # Plot ground truth on x_t axes
         RTCDebugVisualizer.plot_waypoints(
             axs_xt[:, 1], prev_chunk_left_over, start_from=0, color="red", label="Ground truth"
@@ -924,6 +928,41 @@ class RTCEvaluator:
         self._rescale_axes(corr_axs)
         self._rescale_axes(x1t_axs)
 
+    def _plot_no_rtc_xt_reference(self, no_rtc_tracked_steps, xt_axs, num_steps):
+        """Plot final no-RTC x_t data as orange dashed line on the RTC chart for comparison.
+
+        Args:
+            no_rtc_tracked_steps: List of DebugStep objects containing no-RTC debug steps
+            xt_axs: Matplotlib axes for x_t plots (array of 6 axes, right column)
+            num_steps: Total number of denoising steps for colormap
+        """
+        debug_steps = no_rtc_tracked_steps
+        if not debug_steps:
+            return
+
+        # Plot only the final x_t step as orange dashed line
+        final_step = debug_steps[-1]
+        logging.info("Plotting final no-RTC x_t step as orange dashed reference")
+
+        if final_step.x_t is not None:
+            x_t_chunk = (
+                final_step.x_t[0].cpu().numpy()
+                if len(final_step.x_t.shape) == 3
+                else final_step.x_t.cpu().numpy()
+            )
+
+            num_dims = min(x_t_chunk.shape[-1], 6)
+            for j in range(num_dims):
+                xt_axs[j].plot(
+                    np.arange(0, x_t_chunk.shape[0]),
+                    x_t_chunk[:, j],
+                    color="orange",
+                    linestyle="--",
+                    alpha=0.7,
+                    linewidth=2,
+                    label="No RTC (final)" if j == 0 else "",
+                )
+
     def _rescale_axes(self, axes):
         """Rescale axes to show all data with proper margins.
 

src/lerobot/policies/rtc/modeling_rtc.py

@@ -217,6 +217,11 @@ class RTCProcessor:
             grad_outputs = err.clone().detach()
             correction = torch.autograd.grad(x1_t, x_t, grad_outputs, retain_graph=False)[0]
 
+        # Explicitly nullify correction after execution horizon to ensure exact match with no-RTC
+        # Create a mask that zeros out correction after execution_horizon
+        correction_mask = weights.clone()  # weights already have zeros after execution_horizon
+        correction = correction * correction_mask
+
         max_guidance_weight = torch.as_tensor(self.rtc_config.max_guidance_weight)
         tau_tensor = torch.as_tensor(tau)
         squared_one_minus_tau = (1 - tau_tensor) ** 2