Fixes Warehouse import

.gitattributes

@@ -0,0 +1 @@
+*.usda filter=lfs diff=lfs merge=lfs -text

omniverse/extension/omni.viplanner/omni/viplanner/collectors/viewpoint_sampling.py

@@ -81,9 +81,9 @@ def sample_viewpoints(self, nbr_viewpoints: int, seed: int = 1) -> torch.Tensor:
             sample_idx_select = torch.randperm(sample_idx.sum())[
                 : min(nbr_samples_per_point, nbr_viewpoints - sample_locations_count)
             ]
-            sample_locations[sample_locations_count : sample_locations_count + sample_idx_select.shape[0]] = (
-                self.terrain_analyser.samples[sample_idx][sample_idx_select, :2]
-            )
+            sample_locations[
+                sample_locations_count : sample_locations_count + sample_idx_select.shape[0]
+            ] = self.terrain_analyser.samples[sample_idx][sample_idx_select, :2]
             sample_locations_count += sample_idx_select.shape[0]
             curr_point_idx += 1
             # reset point index if all points are sampled
@@ -239,7 +239,7 @@ def render_viewpoints(self, samples: torch.Tensor):
                             os.path.join(
                                 filedir,
                                 "semantics" if annotator == "semantic_segmentation" else "rgb",
-                                f"{image_idx[cam_idx]}".zfill(4) + ".png"
+                                f"{image_idx[cam_idx]}".zfill(4) + ".png",
                             ),
                             cv2.cvtColor(output.astype(np.uint8), cv2.COLOR_RGB2BGR),
                         )

omniverse/extension/omni.viplanner/omni/viplanner/config/matterport_cfg.py

@@ -11,7 +11,7 @@
 from omni.isaac.lab.managers import ObservationTermCfg as ObsTerm
 from omni.isaac.lab.managers import SceneEntityCfg
 from omni.isaac.lab.scene import InteractiveSceneCfg
-from omni.isaac.lab.sensors import ContactSensorCfg, patterns, CameraCfg
+from omni.isaac.lab.sensors import CameraCfg, ContactSensorCfg, patterns
 from omni.isaac.lab.utils import configclass
 from omni.isaac.matterport.config import MatterportImporterCfg
 from omni.isaac.matterport.domains import MatterportRayCasterCfg
@@ -121,7 +121,7 @@ class TerrainSceneCfg(InteractiveSceneCfg):
         debug_vis=False,
         mesh_prim_paths=["${USER_PATH_TO_USD}/matterport.ply"],
     )
-    
+
     # NOTE: comment the following lines to only render the semantic segmentation and depth images
     rgb_camera = CameraCfg(
         prim_path="{ENV_REGEX_NS}/Robot/base/rgb_camera",

omniverse/extension/omni.viplanner/omni/viplanner/config/warehouse_cfg.py

@@ -41,7 +41,7 @@ class TerrainSceneCfg(InteractiveSceneCfg):
             static_friction=1.0,
             dynamic_friction=1.0,
         ),
-        usd_path=os.path.join(DATA_DIR, "warehouse", "warehouse_new.usd"),
+        usd_path=os.path.join(DATA_DIR, "warehouse", "warehouse_new.usda"),
         groundplane=True,
         sem_mesh_to_class_map=os.path.join(DATA_DIR, "warehouse", "keyword_mapping.yml"),
         people_config_file=os.path.join(DATA_DIR, "warehouse", "people_cfg.yml"),

viplanner/utils/dataset.py

@@ -621,7 +621,7 @@ def filter_obs_inflation(self) -> None:
         print("DONE!")
         print(
             "[INFO] odom points outside obs inflation :"
-            f" \t{self.nb_odom_points} ({round(self.nb_odom_points/nb_odom_point_prev*100, 2)} %)"
+            f" \t{self.nb_odom_points} ({round(self.nb_odom_points / nb_odom_point_prev * 100, 2)} %)"
         )
 
         return
@@ -1072,16 +1072,16 @@ def split_samples(
             f"[INFO] datamix containing {odom.shape[0]} suitable odom-goal"
             " pairs: \n"
             "\t fov               :"
-            f" \t{int(odom.shape[0] * ratio_fov_samples)  } ({round(ratio_fov_samples*100, 2)} %) \n"
+            f" \t{int(odom.shape[0] * ratio_fov_samples)} ({round(ratio_fov_samples * 100, 2)} %) \n"
             "\t front of robot    :"
-            f" \t{int(odom.shape[0] * ratio_front_samples)} ({round(ratio_front_samples*100, 2)} %) \n"
+            f" \t{int(odom.shape[0] * ratio_front_samples)} ({round(ratio_front_samples * 100, 2)} %) \n"
             "\t back of robot     :"
-            f" \t{int(odom.shape[0] * ratio_back_samples) } ({round(ratio_back_samples*100, 2)} %) \n"
+            f" \t{int(odom.shape[0] * ratio_back_samples)} ({round(ratio_back_samples * 100, 2)} %) \n"
             "from"
-            f" {self.odom_used} ({round(self.odom_used/self.nb_odom_points*100, 2)} %)"
+            f" {self.odom_used} ({round(self.odom_used / self.nb_odom_points * 100, 2)} %)"
             " different starting points where \n"
             "\t non-suitable filter:"
-            f" {self.odom_no_suitable_goals} ({round(self.odom_no_suitable_goals/self.nb_odom_points*100, 2)} %)"
+            f" {self.odom_no_suitable_goals} ({round(self.odom_no_suitable_goals / self.nb_odom_points * 100, 2)} %)"
         )
 
         # generate split

viplanner/utils/eval_utils.py

@@ -211,7 +211,7 @@ def plt_single_model(self, eval_dir: str, show: bool = True) -> None:
         ax.set_title(
             (
                 "Avg increase of path length is"
-                f" {round(np.mean(self.path_extension), 5)*100:.2f}% for"
+                f" {round(np.mean(self.path_extension), 5) * 100:.2f}% for"
                 " successful paths with tolerance of"
                 f" {self.distance_tolerance}"
             ),
@@ -255,7 +255,7 @@ def plt_single_model(self, eval_dir: str, show: bool = True) -> None:
         ax1.set_title(
             (
                 f"With a tolerance of {self.distance_tolerance} are"
-                f" {round(goal_success_mean, 5)*100:.2f} % of goals reached"
+                f" {round(goal_success_mean, 5) * 100:.2f} % of goals reached"
             ),
             fontsize=16,
         )
@@ -416,7 +416,7 @@ def plt_comparison(
                 unique_goal_length + bar_pos,
                 mean_path_extension,
                 width=bar_width,
-                label=(f"{model_name} (avg {round(avg_increase, 5)*100:.2f} %))"),
+                label=(f"{model_name} (avg {round(avg_increase, 5) * 100:.2f} %))"),
                 alpha=0.8,
             )  # yerr=std_path_extension,
             # axs_path.plot(goal_length_path_exists, mean_path_extension, label=f'{model_name} ({round(avg_increase, 5)*100:.2f} %))')
@@ -428,7 +428,7 @@ def plt_comparison(
                 unique_goal_length + bar_pos,
                 mean_goal_distance,
                 width=bar_width,
-                label=(f"{model_name} (success rate" f" {round(goal_success, 5)*100:.2f} %)"),
+                label=(f"{model_name} (success rate" f" {round(goal_success, 5) * 100:.2f} %)"),
                 alpha=0.8,
             )  # yerr=std_goal_distance,
             # axs_goal.plot(unique_goal_length, mean_goal_distance, label=f'{model_name} ({round(goal_success, 5)*100:.2f} %)')

viplanner/utils/trainer.py

@@ -495,8 +495,8 @@ def _train_epoch(
             train_loss += loss.item()
             enumerater.set_description(
                 f"Epoch: {epoch} in Env: "
-                f"({env_id+1}/{len(self._cfg.env_list)-1}) "
-                f"- train loss:{round(train_loss/(batch_idx+1), 4)} on"
+                f"({env_id + 1} / {len(self._cfg.env_list) - 1}) "
+                f"- train loss:{round(train_loss / (batch_idx + 1), 4)} on"
                 f" {batch_idx}/{batches}"
             )
         return train_loss / (batch_idx + 1)