diff --git a/experiments/pretrained_checkpoints/urbandriver_checkpoint.ckpt b/experiments/pretrained_checkpoints/urbandriver_checkpoint.ckpt
index 0f6db3cfe1af1c5d991ab7dd558a83ac33009e05..7f72d0e465b10d094390fb181f82c3aba35f44b2 100644
Binary files a/experiments/pretrained_checkpoints/urbandriver_checkpoint.ckpt and b/experiments/pretrained_checkpoints/urbandriver_checkpoint.ckpt differ
diff --git a/experiments/relavance_construction.ipynb b/experiments/relavance_construction.ipynb
index 802f941a62b42fd592d70c46144d83e756d5b12c..d0c83dfbc069f68826c715ef087ce4d7dcf899a4 100644
--- a/experiments/relavance_construction.ipynb
+++ b/experiments/relavance_construction.ipynb
@@ -4,7 +4,9 @@
"cell_type": "code",
"execution_count": 1,
"id": "3e31b233",
- "metadata": {},
+ "metadata": {
+ "scrolled": true
+ },
"outputs": [],
"source": [
"import asyncio\n",
@@ -19,7 +21,9 @@
"cell_type": "code",
"execution_count": 2,
"id": "5a4f8ab4",
- "metadata": {},
+ "metadata": {
+ "scrolled": true
+ },
"outputs": [],
"source": [
"import os\n",
@@ -34,7 +38,9 @@
"cell_type": "code",
"execution_count": 3,
"id": "64d027e9",
- "metadata": {},
+ "metadata": {
+ "scrolled": true
+ },
"outputs": [],
"source": [
"import hydra\n",
@@ -71,12 +77,14 @@
"cell_type": "code",
"execution_count": 4,
"id": "6f7bd297",
- "metadata": {},
+ "metadata": {
+ "scrolled": true
+ },
"outputs": [
{
"data": {
"text/plain": [
- "'/host_home/nuplan-devkit/henry_experiments'"
+ "'/host_home/Repos/nuplan-devkit/experiments'"
]
},
"execution_count": 4,
@@ -90,52 +98,66 @@
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": 22,
"id": "ec762c52",
- "metadata": {},
- "outputs": [],
+ "metadata": {
+ "scrolled": true
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "default_simulation\n"
+ ]
+ }
+ ],
"source": [
- "# Location of path with all simulation configs\n",
- "SAVE_DIR = Path('/home/ehdykhne/nuplan-devkit/henry_experiments') / 'relavance_construction' # optionally replace with persistent dir\n",
+ "from tutorials.utils.tutorial_utils import construct_simulation_hydra_paths\n",
"\n",
+ "# Location of paths with all simulation configs\n",
+ "BASE_CONFIG_PATH = os.path.join(os.getenv('NUPLAN_TUTORIAL_PATH', ''), '../nuplan/planning/script')\n",
+ "simulation_hydra_paths = construct_simulation_hydra_paths(BASE_CONFIG_PATH)\n",
"\n",
- "CONFIG_PATH = '../nuplan/planning/script/config/simulation'\n",
- "CONFIG_NAME = 'default_simulation'\n",
- "\n",
- "CHECKPOINT_PATH='pretrained_checkpoints/gc_pgp_checkpoint.ckpt'\n",
+ "# Create a temporary directory to store the simulation artifacts\n",
"\n",
- "# Select the planner and simulation challenge\n",
- "PLANNER = 'ml_planner' # [simple_planner, ml_planner]\n",
- "CHALLENGE = 'closed_loop_reactive_agents' # [open_loop_boxes, closed_loop_nonreactive_agents, closed_loop_reactive_agents]\n",
"DATASET_PARAMS = [\n",
- " 'scenario_builder=nuplan_mini', # use nuplan mini database\n",
+ " 'scenario_builder=nuplan_mini', # use nuplan mini database (2.5h of 8 autolabeled logs in Las Vegas)\n",
" 'scenario_filter=all_scenarios', # initially select all scenarios in the database\n",
" 'scenario_filter.scenario_types=[near_multiple_vehicles, on_pickup_dropoff, starting_unprotected_cross_turn, high_magnitude_jerk]', # select scenario types\n",
- " 'scenario_filter.num_scenarios_per_type=2', # use 2 scenarios per scenario type\n",
+ " 'scenario_filter.num_scenarios_per_type=5', # use 5 scenarios per scenario type\n",
+ "# 'scenario_filter=one_continuous_log', # simulate only one log\n",
+ "# \"scenario_filter.log_names=['2021.06.14.16.48.02_veh-12_04057_04438']\",\n",
+ "# 'scenario_filter.limit_total_scenarios=1', # use 1 total scenarios\n",
"]\n",
- "\n",
- "# Name of the experiment\n",
- "EXPERIMENT = 'simulation_simple_experiment'\n",
- "\n",
+ "ckpt_dir = '/home/ehdykhne/Repos/nuplan-devkit/experiments/pretrained_checkpoints/pdm_offset_checkpoint.ckpt'\n",
+ "#'/home/sacardoz/checkpoints/urbandriver_checkpoint.ckpt'\n",
+ "#\"/home/sacardoz/tutorial_vector_framework/training_simple_vector_experiment/train_default_simple_vector/2023.11.23.09.55.21/best_model/epoch.ckpt\"\n",
+ "#\"/home/sacardoz/training_raster_experiment/train_default_raster/2023.11.23.07.36.36/best_model/epoch.ckpt\"\n",
"# Initialize configuration management system\n",
"hydra.core.global_hydra.GlobalHydra.instance().clear() # reinitialize hydra if already initialized\n",
- "hydra.initialize(config_path=CONFIG_PATH)\n",
+ "hydra.initialize(config_path=simulation_hydra_paths.config_path)\n",
"\n",
"# Compose the configuration\n",
- "cfg = hydra.compose(config_name=CONFIG_NAME, overrides=[\n",
- " f'experiment_name={EXPERIMENT}',\n",
- " f'planner={PLANNER}',\n",
- " f'model=gc_pgp_model',\n",
- " 'planner.ml_planner.model_config=${model}', # hydra notation to select model config\n",
- " f'planner.ml_planner.checkpoint_path={CHECKPOINT_PATH}', # this path can be replaced by the checkpoint of the model trained in the previous section\n",
- "# f'observation={OBSERVATION}',\n",
- "# 'observation.model_config=${model}',\n",
- "# f'observation.checkpoint_path={CHECKPOINT_PATH}',\n",
- " f'group={SAVE_DIR}',\n",
- " f'+simulation={CHALLENGE}',\n",
+ "print(simulation_hydra_paths.config_name)\n",
+ "cfg = hydra.compose(config_name=simulation_hydra_paths.config_name, overrides=[\n",
+ " '+simulation=closed_loop_reactive_agents',\n",
+ " #'model=pgm_hybrid_model',\n",
+ " 'planner=pdm_hybrid_planner',\n",
+ " f\"planner.pdm_hybrid_planner.checkpoint_path={ckpt_dir}\" ,\n",
+ " #'planner.ml_planner.model_config=${model}',\n",
+ " #f'planner.ml_planner.checkpoint_path={ckpt_dir}',\n",
+ " #f'observation=idm_agents_observation',\n",
+ " #'observation.model_config=${model}',\n",
+ " #f'observation.checkpoint_path={ckpt_dir}',\n",
+ " 'worker=sequential',\n",
+ " '+occlusion=true',\n",
+ " '+occlusion.manager_type=wedge', #options: [range, shadow, wedge]\n",
+ " \"hydra.searchpath=[pkg://tuplan_garage.planning.script.config.common, pkg://tuplan_garage.planning.script.config.simulation, pkg://nuplan.planning.script.config.common, pkg://nuplan.planning.script.experiments]\",\n",
" *DATASET_PARAMS,\n",
- " 'hydra.searchpath=[pkg://tuplan_garage.planning.script.config.common, pkg://tuplan_garage.planning.script.config.simulation, pkg://nuplan.planning.script.config.common, pkg://nuplan.planning.script.experiments]'\n",
- "])"
+ "])\n",
+ "\n",
+ "output_folder = cfg.output_dir"
]
},
{
@@ -156,9 +178,11 @@
},
{
"cell_type": "code",
- "execution_count": 25,
+ "execution_count": 23,
"id": "a1e5ba49",
- "metadata": {},
+ "metadata": {
+ "scrolled": true
+ },
"outputs": [],
"source": [
"from __future__ import annotations\n",
@@ -306,9 +330,11 @@
},
{
"cell_type": "code",
- "execution_count": 26,
+ "execution_count": 24,
"id": "69c6768b",
- "metadata": {},
+ "metadata": {
+ "scrolled": true
+ },
"outputs": [],
"source": [
"import concurrent.futures\n",
@@ -459,9 +485,11 @@
},
{
"cell_type": "code",
- "execution_count": 27,
+ "execution_count": 25,
"id": "69db270c",
- "metadata": {},
+ "metadata": {
+ "scrolled": true
+ },
"outputs": [],
"source": [
"import logging\n",
@@ -637,9 +665,11 @@
},
{
"cell_type": "code",
- "execution_count": 28,
+ "execution_count": 26,
"id": "cc8aa1d5",
- "metadata": {},
+ "metadata": {
+ "scrolled": true
+ },
"outputs": [],
"source": [
"import logging\n",
@@ -653,6 +683,7 @@
"from nuplan.planning.scenario_builder.nuplan_db.nuplan_scenario_builder import NuPlanScenarioBuilder\n",
"from nuplan.planning.script.builders.metric_builder import build_metrics_engines\n",
"from nuplan.planning.script.builders.observation_builder import build_observations\n",
+ "from nuplan.planning.script.builders.occlusion_manager_builder import build_occlusion_manager\n",
"from nuplan.planning.script.builders.planner_builder import build_planners\n",
"from nuplan.planning.script.builders.utils.utils_type import is_target_type\n",
"from nuplan.planning.simulation.callback.abstract_callback import AbstractCallback\n",
@@ -747,6 +778,12 @@
" # Perception\n",
" observations: AbstractObservation = build_observations(cfg.observation, scenario=scenario)\n",
"\n",
+ " # Occlusions\n",
+ " if 'occlusion' in cfg.keys() and cfg.occlusion:\n",
+ " occlusion_manager: AbstractOcclusionManager = build_occlusion_manager(cfg.occlusion, scenario=scenario)\n",
+ " else:\n",
+ " occlusion_manager = None\n",
+ "\n",
" # Metric Engine\n",
" metric_engine = metric_engines_map.get(scenario.scenario_type, None)\n",
" if metric_engine is not None:\n",
@@ -764,6 +801,7 @@
" time_controller=simulation_time_controller,\n",
" observations=observations,\n",
" ego_controller=ego_controller,\n",
+ " occlusion_manager=occlusion_manager,\n",
" scenario=scenario,\n",
" )\n",
"\n",
@@ -788,9 +826,11 @@
},
{
"cell_type": "code",
- "execution_count": 29,
+ "execution_count": 27,
"id": "07f1f668",
- "metadata": {},
+ "metadata": {
+ "scrolled": true
+ },
"outputs": [],
"source": [
"import logging\n",
@@ -921,9 +961,11 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 28,
"id": "5c182d7e",
- "metadata": {},
+ "metadata": {
+ "scrolled": false
+ },
"outputs": [
{
"name": "stderr",
@@ -936,235 +978,672 @@
"name": "stdout",
"output_type": "stream",
"text": [
- "hohohohohohohohohohohoh\n",
- "2023-11-21 08:43:55,588 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/main_callback_builder.py:20} Building MultiMainCallback...\n",
- "2023-11-21 08:43:55,588 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/main_callback_builder.py:20} Building MultiMainCallback...\n",
- "2023-11-21 08:43:55,640 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/main_callback_builder.py:35} Building MultiMainCallback: 4...DONE!\n",
- "2023-11-21 08:43:55,640 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/main_callback_builder.py:35} Building MultiMainCallback: 4...DONE!\n",
- "2023-11-21 08:43:56,504 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:19} Building WorkerPool...\n",
- "2023-11-21 08:43:56,504 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:19} Building WorkerPool...\n",
- "2023-11-21 08:43:56,513 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/utils/multithreading/worker_ray.py:78} Starting ray local!\n",
- "2023-11-21 08:43:56,513 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/utils/multithreading/worker_ray.py:78} Starting ray local!\n"
+ "2023-12-06 02:53:02,337 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/main_callback_builder.py:20} Building MultiMainCallback...\n",
+ "2023-12-06 02:53:02,337 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/main_callback_builder.py:20} Building MultiMainCallback...\n",
+ "2023-12-06 02:53:02,362 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/main_callback_builder.py:35} Building MultiMainCallback: 4...DONE!\n",
+ "2023-12-06 02:53:02,362 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/main_callback_builder.py:35} Building MultiMainCallback: 4...DONE!\n",
+ "2023-12-06 02:53:02,532 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:19} Building WorkerPool...\n",
+ "2023-12-06 02:53:02,532 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:19} Building WorkerPool...\n",
+ "2023-12-06 02:53:02,533 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:101} Worker: Sequential\n",
+ "2023-12-06 02:53:02,533 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:101} Worker: Sequential\n",
+ "2023-12-06 02:53:02,533 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:102} Number of nodes: 1\n",
+ "Number of CPUs per node: 1\n",
+ "Number of GPUs per node: 0\n",
+ "Number of threads across all nodes: 1\n",
+ "2023-12-06 02:53:02,533 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:102} Number of nodes: 1\n",
+ "Number of CPUs per node: 1\n",
+ "Number of GPUs per node: 0\n",
+ "Number of threads across all nodes: 1\n",
+ "2023-12-06 02:53:02,533 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:27} Building WorkerPool...DONE!\n",
+ "2023-12-06 02:53:02,533 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:27} Building WorkerPool...DONE!\n",
+ "2023-12-06 02:53:02,533 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:32} Building experiment folders...\n",
+ "2023-12-06 02:53:02,533 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:32} Building experiment folders...\n",
+ "2023-12-06 02:53:02,533 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:35} \n",
+ "\n",
+ "\tFolder where all results are stored: /home/ehdykhne/Repos/Datasets/nuplan/exp/exp/simulation/closed_loop_reactive_agents/2023.12.06.02.52.57\n",
+ "\n",
+ "2023-12-06 02:53:02,533 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:35} \n",
+ "\n",
+ "\tFolder where all results are stored: /home/ehdykhne/Repos/Datasets/nuplan/exp/exp/simulation/closed_loop_reactive_agents/2023.12.06.02.52.57\n",
+ "\n",
+ "2023-12-06 02:53:02,534 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:70} Building experiment folders...DONE!\n",
+ "2023-12-06 02:53:02,534 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:70} Building experiment folders...DONE!\n",
+ "2023-12-06 02:53:02,535 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:52} Building AbstractCallback...\n",
+ "2023-12-06 02:53:02,535 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:52} Building AbstractCallback...\n",
+ "2023-12-06 02:53:02,535 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:68} Building AbstractCallback: 0...DONE!\n",
+ "2023-12-06 02:53:02,535 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:68} Building AbstractCallback: 0...DONE!\n",
+ "2023-12-06 02:53:02,535 INFO {/tmp/ipykernel_4075/1187658140.py:48} Building simulations...\n",
+ "2023-12-06 02:53:02,535 INFO {/tmp/ipykernel_4075/1187658140.py:48} Building simulations...\n",
+ "2023-12-06 02:53:02,535 INFO {/tmp/ipykernel_4075/1187658140.py:54} Extracting scenarios...\n",
+ "2023-12-06 02:53:02,535 INFO {/tmp/ipykernel_4075/1187658140.py:54} Extracting scenarios...\n",
+ "2023-12-06 02:53:02,535 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/common/utils/distributed_scenario_filter.py:83} Building Scenarios in mode DistributedMode.SINGLE_NODE\n",
+ "2023-12-06 02:53:02,535 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/common/utils/distributed_scenario_filter.py:83} Building Scenarios in mode DistributedMode.SINGLE_NODE\n",
+ "2023-12-06 02:53:02,535 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:18} Building AbstractScenarioBuilder...\n",
+ "2023-12-06 02:53:02,535 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:18} Building AbstractScenarioBuilder...\n",
+ "2023-12-06 02:53:02,546 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:21} Building AbstractScenarioBuilder...DONE!\n",
+ "2023-12-06 02:53:02,546 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:21} Building AbstractScenarioBuilder...DONE!\n",
+ "2023-12-06 02:53:02,546 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:35} Building ScenarioFilter...\n",
+ "2023-12-06 02:53:02,546 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:35} Building ScenarioFilter...\n",
+ "2023-12-06 02:53:02,548 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:44} Building ScenarioFilter...DONE!\n",
+ "2023-12-06 02:53:02,548 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:44} Building ScenarioFilter...DONE!\n",
+ "2023-12-06 02:53:03,237 INFO {/tmp/ipykernel_4075/1187658140.py:75} Building metric engines...\n",
+ "2023-12-06 02:53:03,237 INFO {/tmp/ipykernel_4075/1187658140.py:75} Building metric engines...\n",
+ "2023-12-06 02:53:03,306 INFO {/tmp/ipykernel_4075/1187658140.py:77} Building metric engines...DONE\n",
+ "2023-12-06 02:53:03,306 INFO {/tmp/ipykernel_4075/1187658140.py:77} Building metric engines...DONE\n",
+ "2023-12-06 02:53:03,307 INFO {/tmp/ipykernel_4075/1187658140.py:81} Building simulations from 19 scenarios...\n",
+ "2023-12-06 02:53:03,307 INFO {/tmp/ipykernel_4075/1187658140.py:81} Building simulations from 19 scenarios...\n",
+ "2023-12-06 02:53:05,123 INFO {/tmp/ipykernel_4075/1187658140.py:141} Building simulations...DONE!\n",
+ "2023-12-06 02:53:05,123 INFO {/tmp/ipykernel_4075/1187658140.py:141} Building simulations...DONE!\n",
+ "2023-12-06 02:53:05,123 INFO {/tmp/ipykernel_4075/1021999762.py:78} Running simulation...\n",
+ "2023-12-06 02:53:05,123 INFO {/tmp/ipykernel_4075/1021999762.py:78} Running simulation...\n",
+ "2023-12-06 02:53:05,123 INFO {/tmp/ipykernel_4075/1874490930.py:138} Executing runners...\n",
+ "2023-12-06 02:53:05,123 INFO {/tmp/ipykernel_4075/1874490930.py:138} Executing runners...\n",
+ "2023-12-06 02:53:05,123 INFO {/tmp/ipykernel_4075/1049651184.py:83} Starting 19 simulations using Sequential!\n",
+ "2023-12-06 02:53:05,123 INFO {/tmp/ipykernel_4075/1049651184.py:83} Starting 19 simulations using Sequential!\n",
+ "about to start all the runners\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ " \r"
]
},
+ {
+ "ename": "KeyboardInterrupt",
+ "evalue": "",
+ "output_type": "error",
+ "traceback": [
+ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+ "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)",
+ "Cell \u001b[0;32mIn[28], line 2\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[38;5;66;03m# Run the simulation loop (real-time visualization not yet supported, see next section for visualization)\u001b[39;00m\n\u001b[0;32m----> 2\u001b[0m \u001b[43mmain\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcfg\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 4\u001b[0m \u001b[38;5;66;03m# Simple simulation folder for visualization in nuBoard\u001b[39;00m\n\u001b[1;32m 5\u001b[0m simple_simulation_folder \u001b[38;5;241m=\u001b[39m cfg\u001b[38;5;241m.\u001b[39moutput_dir\n",
+ "File \u001b[0;32m/opt/conda/lib/python3.9/site-packages/hydra/main.py:44\u001b[0m, in \u001b[0;36mmain.<locals>.main_decorator.<locals>.decorated_main\u001b[0;34m(cfg_passthrough)\u001b[0m\n\u001b[1;32m 41\u001b[0m \u001b[38;5;129m@functools\u001b[39m\u001b[38;5;241m.\u001b[39mwraps(task_function)\n\u001b[1;32m 42\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mdecorated_main\u001b[39m(cfg_passthrough: Optional[DictConfig] \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m Any:\n\u001b[1;32m 43\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m cfg_passthrough \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m---> 44\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mtask_function\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcfg_passthrough\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 45\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m 46\u001b[0m args \u001b[38;5;241m=\u001b[39m get_args_parser()\n",
+ "Cell \u001b[0;32mIn[27], line 109\u001b[0m, in \u001b[0;36mmain\u001b[0;34m(cfg)\u001b[0m\n\u001b[1;32m 107\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m cfg\u001b[38;5;241m.\u001b[39msimulation_log_main_path \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mSimulation_log_main_path must not be set when running simulation.\u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[1;32m 108\u001b[0m \u001b[38;5;66;03m# Execute simulation with preconfigured planner(s).\u001b[39;00m\n\u001b[0;32m--> 109\u001b[0m \u001b[43mrun_simulation_main\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcfg\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcfg\u001b[49m\u001b[43m)\u001b[49m\n",
+ "Cell \u001b[0;32mIn[27], line 79\u001b[0m, in \u001b[0;36mrun_simulation_main\u001b[0;34m(cfg, planners)\u001b[0m\n\u001b[1;32m 76\u001b[0m common_builder\u001b[38;5;241m.\u001b[39mprofiler\u001b[38;5;241m.\u001b[39msave_profiler(profiler_name)\n\u001b[1;32m 78\u001b[0m logger\u001b[38;5;241m.\u001b[39minfo(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mRunning simulation...\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m---> 79\u001b[0m \u001b[43mrun_runners\u001b[49m\u001b[43m(\u001b[49m\u001b[43mrunners\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mrunners\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcommon_builder\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcommon_builder\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcfg\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcfg\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mprofiler_name\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mrunning_simulation\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m)\u001b[49m\n\u001b[1;32m 80\u001b[0m logger\u001b[38;5;241m.\u001b[39minfo(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mFinished running simulation!\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
+ "Cell \u001b[0;32mIn[25], line 139\u001b[0m, in \u001b[0;36mrun_runners\u001b[0;34m(runners, common_builder, profiler_name, cfg)\u001b[0m\n\u001b[1;32m 136\u001b[0m common_builder\u001b[38;5;241m.\u001b[39mprofiler\u001b[38;5;241m.\u001b[39mstart_profiler(profiler_name)\n\u001b[1;32m 138\u001b[0m logger\u001b[38;5;241m.\u001b[39minfo(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mExecuting runners...\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m--> 139\u001b[0m reports \u001b[38;5;241m=\u001b[39m \u001b[43mexecute_runners\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m 140\u001b[0m \u001b[43m \u001b[49m\u001b[43mrunners\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mrunners\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 141\u001b[0m \u001b[43m \u001b[49m\u001b[43mworker\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcommon_builder\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mworker\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 142\u001b[0m \u001b[43m \u001b[49m\u001b[43mnum_gpus\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcfg\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mnumber_of_gpus_allocated_per_simulation\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 143\u001b[0m \u001b[43m \u001b[49m\u001b[43mnum_cpus\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcfg\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mnumber_of_cpus_allocated_per_simulation\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 144\u001b[0m \u001b[43m \u001b[49m\u001b[43mexit_on_failure\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcfg\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mexit_on_failure\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 145\u001b[0m \u001b[43m \u001b[49m\u001b[43mverbose\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcfg\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mverbose\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 146\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 147\u001b[0m logger\u001b[38;5;241m.\u001b[39minfo(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mFinished executing runners!\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[1;32m 149\u001b[0m \u001b[38;5;66;03m# Save RunnerReports as parquet file\u001b[39;00m\n",
+ "Cell \u001b[0;32mIn[24], line 85\u001b[0m, in \u001b[0;36mexecute_runners\u001b[0;34m(runners, worker, num_gpus, num_cpus, exit_on_failure, verbose)\u001b[0m\n\u001b[1;32m 83\u001b[0m logger\u001b[38;5;241m.\u001b[39minfo(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mStarting \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mnumber_of_sims\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m simulations using \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mworker\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__class__\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m!\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m 84\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mabout to start all the runners\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m---> 85\u001b[0m reports: List[RunnerReport] \u001b[38;5;241m=\u001b[39m \u001b[43mworker\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmap\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m 86\u001b[0m \u001b[43m \u001b[49m\u001b[43mTask\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfn\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mrun_simulation\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mnum_gpus\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mnum_gpus\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mnum_cpus\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mnum_cpus\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mrunners\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mexit_on_failure\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mverbose\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mverbose\u001b[49m\n\u001b[1;32m 87\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 88\u001b[0m \u001b[38;5;66;03m# Store the results in a dictionary so we can easily store error tracebacks in the next step, if needed\u001b[39;00m\n\u001b[1;32m 89\u001b[0m results: Dict[Tuple[\u001b[38;5;28mstr\u001b[39m, \u001b[38;5;28mstr\u001b[39m, \u001b[38;5;28mstr\u001b[39m], RunnerReport] \u001b[38;5;241m=\u001b[39m {\n\u001b[1;32m 90\u001b[0m (report\u001b[38;5;241m.\u001b[39mscenario_name, report\u001b[38;5;241m.\u001b[39mplanner_name, report\u001b[38;5;241m.\u001b[39mlog_name): report \u001b[38;5;28;01mfor\u001b[39;00m report \u001b[38;5;129;01min\u001b[39;00m reports\n\u001b[1;32m 91\u001b[0m }\n",
+ "File \u001b[0;32m~/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:117\u001b[0m, in \u001b[0;36mWorkerPool.map\u001b[0;34m(self, task, verbose, *item_lists)\u001b[0m\n\u001b[1;32m 115\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m verbose:\n\u001b[1;32m 116\u001b[0m logger\u001b[38;5;241m.\u001b[39minfo(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mSubmitting \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mmax_size\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m tasks!\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m--> 117\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_map\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtask\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43maligned_item_lists\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mverbose\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mverbose\u001b[49m\u001b[43m)\u001b[49m\n",
+ "File \u001b[0;32m~/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_sequential.py:32\u001b[0m, in \u001b[0;36mSequential._map\u001b[0;34m(self, task, verbose, *item_lists)\u001b[0m\n\u001b[1;32m 30\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m task\u001b[38;5;241m.\u001b[39mnum_cpus \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m [\u001b[38;5;28;01mNone\u001b[39;00m, \u001b[38;5;241m1\u001b[39m]:\n\u001b[1;32m 31\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mExpected num_cpus to be 1 or unset for Sequential worker, got \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mtask\u001b[38;5;241m.\u001b[39mnum_cpus\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m---> 32\u001b[0m output \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m 33\u001b[0m task\u001b[38;5;241m.\u001b[39mfn(\u001b[38;5;241m*\u001b[39margs)\n\u001b[1;32m 34\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m args \u001b[38;5;129;01min\u001b[39;00m tqdm(\n\u001b[1;32m 35\u001b[0m \u001b[38;5;28mzip\u001b[39m(\u001b[38;5;241m*\u001b[39mitem_lists),\n\u001b[1;32m 36\u001b[0m leave\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mFalse\u001b[39;00m,\n\u001b[1;32m 37\u001b[0m total\u001b[38;5;241m=\u001b[39mget_max_size_of_arguments(\u001b[38;5;241m*\u001b[39mitem_lists),\n\u001b[1;32m 38\u001b[0m desc\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mSequential\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[1;32m 39\u001b[0m disable\u001b[38;5;241m=\u001b[39m\u001b[38;5;129;01mnot\u001b[39;00m verbose,\n\u001b[1;32m 40\u001b[0m )\n\u001b[1;32m 41\u001b[0m ]\n\u001b[1;32m 42\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m output\n",
+ "File \u001b[0;32m~/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_sequential.py:33\u001b[0m, in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 30\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m task\u001b[38;5;241m.\u001b[39mnum_cpus \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m [\u001b[38;5;28;01mNone\u001b[39;00m, \u001b[38;5;241m1\u001b[39m]:\n\u001b[1;32m 31\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mExpected num_cpus to be 1 or unset for Sequential worker, got \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mtask\u001b[38;5;241m.\u001b[39mnum_cpus\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m'\u001b[39m)\n\u001b[1;32m 32\u001b[0m output \u001b[38;5;241m=\u001b[39m [\n\u001b[0;32m---> 33\u001b[0m \u001b[43mtask\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mfn\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 34\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m args \u001b[38;5;129;01min\u001b[39;00m tqdm(\n\u001b[1;32m 35\u001b[0m \u001b[38;5;28mzip\u001b[39m(\u001b[38;5;241m*\u001b[39mitem_lists),\n\u001b[1;32m 36\u001b[0m leave\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mFalse\u001b[39;00m,\n\u001b[1;32m 37\u001b[0m total\u001b[38;5;241m=\u001b[39mget_max_size_of_arguments(\u001b[38;5;241m*\u001b[39mitem_lists),\n\u001b[1;32m 38\u001b[0m desc\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mSequential\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[1;32m 39\u001b[0m disable\u001b[38;5;241m=\u001b[39m\u001b[38;5;129;01mnot\u001b[39;00m verbose,\n\u001b[1;32m 40\u001b[0m )\n\u001b[1;32m 41\u001b[0m ]\n\u001b[1;32m 42\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m output\n",
+ "Cell \u001b[0;32mIn[24], line 28\u001b[0m, in \u001b[0;36mrun_simulation\u001b[0;34m(sim_runner, exit_on_failure)\u001b[0m\n\u001b[1;32m 25\u001b[0m start_time \u001b[38;5;241m=\u001b[39m time\u001b[38;5;241m.\u001b[39mperf_counter()\n\u001b[1;32m 27\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m---> 28\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43msim_runner\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 29\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m 30\u001b[0m error \u001b[38;5;241m=\u001b[39m traceback\u001b[38;5;241m.\u001b[39mformat_exc()\n",
+ "Cell \u001b[0;32mIn[23], line 113\u001b[0m, in \u001b[0;36mSimulationRunner.run\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 110\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_simulation\u001b[38;5;241m.\u001b[39mcallback\u001b[38;5;241m.\u001b[39mon_planner_start(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39msimulation\u001b[38;5;241m.\u001b[39msetup, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mplanner)\n\u001b[1;32m 112\u001b[0m \u001b[38;5;66;03m# Plan path based on all planner's inputs\u001b[39;00m\n\u001b[0;32m--> 113\u001b[0m trajectory \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mplanner\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcompute_trajectory\u001b[49m\u001b[43m(\u001b[49m\u001b[43mplanner_input\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 115\u001b[0m \u001b[38;5;66;03m# Propagate simulation based on planner trajectory\u001b[39;00m\n\u001b[1;32m 116\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_simulation\u001b[38;5;241m.\u001b[39mcallback\u001b[38;5;241m.\u001b[39mon_planner_end(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39msimulation\u001b[38;5;241m.\u001b[39msetup, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mplanner, trajectory)\n",
+ "File \u001b[0;32m~/Repos/nuplan-devkit/nuplan/planning/simulation/planner/abstract_planner.py:105\u001b[0m, in \u001b[0;36mAbstractPlanner.compute_trajectory\u001b[0;34m(self, current_input)\u001b[0m\n\u001b[1;32m 103\u001b[0m \u001b[38;5;66;03m# If it raises an exception, catch to record the time then re-raise it.\u001b[39;00m\n\u001b[1;32m 104\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 105\u001b[0m trajectory \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcompute_planner_trajectory\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcurrent_input\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 106\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m 107\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_compute_trajectory_runtimes\u001b[38;5;241m.\u001b[39mappend(time\u001b[38;5;241m.\u001b[39mperf_counter() \u001b[38;5;241m-\u001b[39m start_time)\n",
+ "File \u001b[0;32m~/Repos/tuplan_garage/tuplan_garage/planning/simulation/planner/pdm_planner/pdm_hybrid_planner.py:136\u001b[0m, in \u001b[0;36mPDMHybridPlanner.compute_planner_trajectory\u001b[0;34m(self, current_input)\u001b[0m\n\u001b[1;32m 133\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_centerline \u001b[38;5;241m=\u001b[39m PDMPath(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_get_discrete_centerline(current_lane))\n\u001b[1;32m 135\u001b[0m \u001b[38;5;66;03m# trajectory of PDM-Closed\u001b[39;00m\n\u001b[0;32m--> 136\u001b[0m closed_loop_trajectory \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_get_closed_loop_trajectory\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcurrent_input\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 137\u001b[0m uncorrected_states \u001b[38;5;241m=\u001b[39m closed_loop_trajectory\u001b[38;5;241m.\u001b[39mget_sampled_trajectory()\n\u001b[1;32m 139\u001b[0m \u001b[38;5;66;03m# trajectory of PDM-Offset\u001b[39;00m\n",
+ "File \u001b[0;32m~/Repos/tuplan_garage/tuplan_garage/planning/simulation/planner/pdm_planner/abstract_pdm_closed_planner.py:187\u001b[0m, in \u001b[0;36mAbstractPDMClosedPlanner._get_closed_loop_trajectory\u001b[0;34m(self, current_input)\u001b[0m\n\u001b[1;32m 185\u001b[0m \u001b[38;5;66;03m# 6.b Otherwise, extend and output best proposal\u001b[39;00m\n\u001b[1;32m 186\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m trajectory \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m--> 187\u001b[0m trajectory \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_generator\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mgenerate_trajectory\u001b[49m\u001b[43m(\u001b[49m\u001b[43mnp\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43margmax\u001b[49m\u001b[43m(\u001b[49m\u001b[43mproposal_scores\u001b[49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 189\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m trajectory\n",
+ "File \u001b[0;32m~/Repos/tuplan_garage/tuplan_garage/planning/simulation/planner/pdm_planner/proposal/pdm_generator.py:135\u001b[0m, in \u001b[0;36mPDMGenerator.generate_trajectory\u001b[0;34m(self, proposal_idx)\u001b[0m\n\u001b[1;32m 132\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_time_point_list\u001b[38;5;241m.\u001b[39mappend(current_time_point)\n\u001b[1;32m 134\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_update_leading_agents(lateral_batch_idcs, time_idx)\n\u001b[0;32m--> 135\u001b[0m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_update_idm_states\u001b[49m\u001b[43m(\u001b[49m\u001b[43mlateral_batch_idcs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtime_idx\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 136\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_update_states_se2(lateral_batch_idcs, time_idx)\n\u001b[1;32m 138\u001b[0m \u001b[38;5;66;03m# convert array representation to list of EgoState class\u001b[39;00m\n",
+ "File \u001b[0;32m~/Repos/tuplan_garage/tuplan_garage/planning/simulation/planner/pdm_planner/proposal/pdm_generator.py:256\u001b[0m, in \u001b[0;36mPDMGenerator._update_idm_states\u001b[0;34m(self, lateral_batch_idcs, time_idx)\u001b[0m\n\u001b[1;32m 251\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m time_idx \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m0\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mPDMGenerator: call _initialize_states first!\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m 252\u001b[0m longitudinal_idcs \u001b[38;5;241m=\u001b[39m [\n\u001b[1;32m 253\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_proposal_manager[proposal_idx]\u001b[38;5;241m.\u001b[39mlongitudinal_idx\n\u001b[1;32m 254\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m proposal_idx \u001b[38;5;129;01min\u001b[39;00m lateral_batch_idcs\n\u001b[1;32m 255\u001b[0m ]\n\u001b[0;32m--> 256\u001b[0m next_idm_states \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_proposal_manager\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mlongitudinal_policies\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpropagate\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m 257\u001b[0m \u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_state_idm_array\u001b[49m\u001b[43m[\u001b[49m\u001b[43mlateral_batch_idcs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtime_idx\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m-\u001b[39;49m\u001b[43m \u001b[49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 258\u001b[0m \u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_leading_agent_array\u001b[49m\u001b[43m[\u001b[49m\u001b[43mlateral_batch_idcs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtime_idx\u001b[49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 259\u001b[0m \u001b[43m \u001b[49m\u001b[43mlongitudinal_idcs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 260\u001b[0m \u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_sample_interval\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m 261\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 262\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_state_idm_array[lateral_batch_idcs, time_idx] \u001b[38;5;241m=\u001b[39m next_idm_states\n",
+ "File \u001b[0;32m~/Repos/tuplan_garage/tuplan_garage/planning/simulation/planner/pdm_planner/proposal/batch_idm_policy.py:161\u001b[0m, in \u001b[0;36mBatchIDMPolicy.propagate\u001b[0;34m(self, previous_idm_states, leading_agent_states, longitudinal_idcs, sampling_time)\u001b[0m\n\u001b[1;32m 153\u001b[0m \u001b[38;5;28;01massert\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(previous_idm_states) \u001b[38;5;241m==\u001b[39m \u001b[38;5;28mlen\u001b[39m(longitudinal_idcs) \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(\n\u001b[1;32m 154\u001b[0m leading_agent_states\n\u001b[1;32m 155\u001b[0m ) \u001b[38;5;241m==\u001b[39m \u001b[38;5;28mlen\u001b[39m(\n\u001b[1;32m 156\u001b[0m longitudinal_idcs\n\u001b[1;32m 157\u001b[0m ), \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mPDMIDMPolicy: propagate function requires equal length of input arguments!\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m 159\u001b[0m \u001b[38;5;66;03m# state variables\u001b[39;00m\n\u001b[1;32m 160\u001b[0m x_agent, v_agent \u001b[38;5;241m=\u001b[39m (\n\u001b[0;32m--> 161\u001b[0m previous_idm_states[:, \u001b[43mStateIDMIndex\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mPROGRESS\u001b[49m],\n\u001b[1;32m 162\u001b[0m previous_idm_states[:, StateIDMIndex\u001b[38;5;241m.\u001b[39mVELOCITY],\n\u001b[1;32m 163\u001b[0m )\n\u001b[1;32m 165\u001b[0m x_lead, v_lead, l_r_lead \u001b[38;5;241m=\u001b[39m (\n\u001b[1;32m 166\u001b[0m leading_agent_states[:, LeadingAgentIndex\u001b[38;5;241m.\u001b[39mPROGRESS],\n\u001b[1;32m 167\u001b[0m leading_agent_states[:, LeadingAgentIndex\u001b[38;5;241m.\u001b[39mVELOCITY],\n\u001b[1;32m 168\u001b[0m leading_agent_states[:, LeadingAgentIndex\u001b[38;5;241m.\u001b[39mLENGTH_REAR],\n\u001b[1;32m 169\u001b[0m )\n\u001b[1;32m 171\u001b[0m \u001b[38;5;66;03m# parameters\u001b[39;00m\n",
+ "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
+ ]
+ }
+ ],
+ "source": [
+ "# Run the simulation loop (real-time visualization not yet supported, see next section for visualization)\n",
+ "main(cfg)\n",
+ "\n",
+ "# Simple simulation folder for visualization in nuBoard\n",
+ "simple_simulation_folder = cfg.output_dir"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "id": "c221b637",
+ "metadata": {
+ "scrolled": false
+ },
+ "outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
- "2023-11-21 08:44:02,855\tINFO worker.py:1673 -- Started a local Ray instance.\n"
+ "Global seed set to 0\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "2023-11-21 08:44:09,456 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:101} Worker: RayDistributed\n",
- "2023-11-21 08:44:09,456 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:101} Worker: RayDistributed\n",
- "2023-11-21 08:44:09,458 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:102} Number of nodes: 1\n",
- "Number of CPUs per node: 32\n",
- "Number of GPUs per node: 1\n",
- "Number of threads across all nodes: 32\n",
- "2023-11-21 08:44:09,458 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:102} Number of nodes: 1\n",
- "Number of CPUs per node: 32\n",
- "Number of GPUs per node: 1\n",
- "Number of threads across all nodes: 32\n",
- "2023-11-21 08:44:09,458 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:27} Building WorkerPool...DONE!\n",
- "2023-11-21 08:44:09,458 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:27} Building WorkerPool...DONE!\n",
- "2023-11-21 08:44:09,458 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:32} Building experiment folders...\n",
- "2023-11-21 08:44:09,458 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:32} Building experiment folders...\n",
- "2023-11-21 08:44:09,459 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:35} \n",
+ "2023-12-06 02:31:37,952 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/main_callback_builder.py:20} Building MultiMainCallback...\n",
+ "2023-12-06 02:31:37,958 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/main_callback_builder.py:35} Building MultiMainCallback: 4...DONE!\n",
+ "2023-12-06 02:31:38,122 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:19} Building WorkerPool...\n",
+ "2023-12-06 02:31:38,122 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:19} Building WorkerPool...\n",
+ "2023-12-06 02:31:38,123 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:101} Worker: Sequential\n",
+ "2023-12-06 02:31:38,123 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:101} Worker: Sequential\n",
+ "2023-12-06 02:31:38,123 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:102} Number of nodes: 1\n",
+ "Number of CPUs per node: 1\n",
+ "Number of GPUs per node: 0\n",
+ "Number of threads across all nodes: 1\n",
+ "2023-12-06 02:31:38,123 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:102} Number of nodes: 1\n",
+ "Number of CPUs per node: 1\n",
+ "Number of GPUs per node: 0\n",
+ "Number of threads across all nodes: 1\n",
+ "2023-12-06 02:31:38,123 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:27} Building WorkerPool...DONE!\n",
+ "2023-12-06 02:31:38,123 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:27} Building WorkerPool...DONE!\n",
+ "2023-12-06 02:31:38,123 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:32} Building experiment folders...\n",
+ "2023-12-06 02:31:38,123 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:32} Building experiment folders...\n",
+ "2023-12-06 02:31:38,123 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:35} \n",
"\n",
- "\tFolder where all results are stored: /home/ehdykhne/nuplan-devkit/henry_experiments/relavance_construction/simulation_simple_experiment/closed_loop_reactive_agents/2023.11.21.08.11.31\n",
+ "\tFolder where all results are stored: /home/ehdykhne/Repos/Datasets/nuplan/exp/exp/simulation/closed_loop_reactive_agents/2023.12.06.02.28.50\n",
"\n",
- "2023-11-21 08:44:09,459 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:35} \n",
+ "2023-12-06 02:31:38,123 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:35} \n",
"\n",
- "\tFolder where all results are stored: /home/ehdykhne/nuplan-devkit/henry_experiments/relavance_construction/simulation_simple_experiment/closed_loop_reactive_agents/2023.11.21.08.11.31\n",
+ "\tFolder where all results are stored: /home/ehdykhne/Repos/Datasets/nuplan/exp/exp/simulation/closed_loop_reactive_agents/2023.12.06.02.28.50\n",
"\n",
- "2023-11-21 08:44:09,475 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:70} Building experiment folders...DONE!\n",
- "2023-11-21 08:44:09,475 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:70} Building experiment folders...DONE!\n",
- "2023-11-21 08:44:09,477 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:52} Building AbstractCallback...\n",
- "2023-11-21 08:44:09,477 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:52} Building AbstractCallback...\n",
- "2023-11-21 08:44:09,477 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:68} Building AbstractCallback: 0...DONE!\n",
- "2023-11-21 08:44:09,477 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:68} Building AbstractCallback: 0...DONE!\n",
- "lalalalalalala\n",
- "2023-11-21 08:44:09,478 INFO {/tmp/ipykernel_68186/2067711137.py:48} Building simulations...\n",
- "2023-11-21 08:44:09,478 INFO {/tmp/ipykernel_68186/2067711137.py:48} Building simulations...\n",
- "2023-11-21 08:44:09,479 INFO {/tmp/ipykernel_68186/2067711137.py:54} Extracting scenarios...\n",
- "2023-11-21 08:44:09,479 INFO {/tmp/ipykernel_68186/2067711137.py:54} Extracting scenarios...\n",
- "2023-11-21 08:44:09,481 INFO {/home/ehdykhne/nuplan-devkit/nuplan/common/utils/distributed_scenario_filter.py:83} Building Scenarios in mode DistributedMode.SINGLE_NODE\n",
- "2023-11-21 08:44:09,481 INFO {/home/ehdykhne/nuplan-devkit/nuplan/common/utils/distributed_scenario_filter.py:83} Building Scenarios in mode DistributedMode.SINGLE_NODE\n",
- "2023-11-21 08:44:09,481 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:18} Building AbstractScenarioBuilder...\n",
- "2023-11-21 08:44:09,481 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:18} Building AbstractScenarioBuilder...\n",
- "2023-11-21 08:44:09,574 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:21} Building AbstractScenarioBuilder...DONE!\n",
- "2023-11-21 08:44:09,574 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:21} Building AbstractScenarioBuilder...DONE!\n",
- "2023-11-21 08:44:09,579 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:35} Building ScenarioFilter...\n",
- "2023-11-21 08:44:09,579 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:35} Building ScenarioFilter...\n",
- "2023-11-21 08:44:09,585 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:44} Building ScenarioFilter...DONE!\n",
- "2023-11-21 08:44:09,585 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:44} Building ScenarioFilter...DONE!\n"
+ "2023-12-06 02:31:38,124 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:70} Building experiment folders...DONE!\n",
+ "2023-12-06 02:31:38,124 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:70} Building experiment folders...DONE!\n",
+ "2023-12-06 02:31:38,124 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:52} Building AbstractCallback...\n",
+ "2023-12-06 02:31:38,124 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:52} Building AbstractCallback...\n",
+ "2023-12-06 02:31:38,124 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:68} Building AbstractCallback: 0...DONE!\n",
+ "2023-12-06 02:31:38,124 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:68} Building AbstractCallback: 0...DONE!\n",
+ "2023-12-06 02:31:38,125 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:49} Building simulations...\n",
+ "2023-12-06 02:31:38,125 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:49} Building simulations...\n",
+ "2023-12-06 02:31:38,125 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:55} Extracting scenarios...\n",
+ "2023-12-06 02:31:38,125 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:55} Extracting scenarios...\n",
+ "2023-12-06 02:31:38,125 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/common/utils/distributed_scenario_filter.py:83} Building Scenarios in mode DistributedMode.SINGLE_NODE\n",
+ "2023-12-06 02:31:38,125 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/common/utils/distributed_scenario_filter.py:83} Building Scenarios in mode DistributedMode.SINGLE_NODE\n",
+ "2023-12-06 02:31:38,125 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:18} Building AbstractScenarioBuilder...\n",
+ "2023-12-06 02:31:38,125 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:18} Building AbstractScenarioBuilder...\n",
+ "2023-12-06 02:31:38,136 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:21} Building AbstractScenarioBuilder...DONE!\n",
+ "2023-12-06 02:31:38,136 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:21} Building AbstractScenarioBuilder...DONE!\n",
+ "2023-12-06 02:31:38,136 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:35} Building ScenarioFilter...\n",
+ "2023-12-06 02:31:38,136 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:35} Building ScenarioFilter...\n",
+ "2023-12-06 02:31:38,137 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:44} Building ScenarioFilter...DONE!\n",
+ "2023-12-06 02:31:38,137 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:44} Building ScenarioFilter...DONE!\n",
+ "2023-12-06 02:31:38,352 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:76} Building metric engines...\n",
+ "2023-12-06 02:31:38,352 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:76} Building metric engines...\n",
+ "2023-12-06 02:31:38,369 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:78} Building metric engines...DONE\n",
+ "2023-12-06 02:31:38,369 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:78} Building metric engines...DONE\n",
+ "2023-12-06 02:31:38,369 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:82} Building simulations from 1 scenarios...\n",
+ "2023-12-06 02:31:38,369 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:82} Building simulations from 1 scenarios...\n",
+ "2023-12-06 02:31:38,756 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:142} Building simulations...DONE!\n",
+ "2023-12-06 02:31:38,756 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:142} Building simulations...DONE!\n",
+ "2023-12-06 02:31:38,756 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/run_simulation.py:114} Running simulation...\n",
+ "2023-12-06 02:31:38,756 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/run_simulation.py:114} Running simulation...\n",
+ "2023-12-06 02:31:38,757 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/utils.py:138} Executing runners...\n",
+ "2023-12-06 02:31:38,757 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/utils.py:138} Executing runners...\n",
+ "2023-12-06 02:31:38,757 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/runner/executor.py:82} Starting 1 simulations using Sequential!\n",
+ "2023-12-06 02:31:38,757 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/runner/executor.py:82} Starting 1 simulations using Sequential!\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
- "Ray objects: 100%|██████████| 32/32 [01:17<00:00, 2.42s/it]\n"
+ " \r"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "2023-11-21 08:45:27,214 INFO {/tmp/ipykernel_68186/2067711137.py:75} Building metric engines...\n",
- "2023-11-21 08:45:27,214 INFO {/tmp/ipykernel_68186/2067711137.py:75} Building metric engines...\n",
- "2023-11-21 08:45:27,576 INFO {/tmp/ipykernel_68186/2067711137.py:77} Building metric engines...DONE\n",
- "2023-11-21 08:45:27,576 INFO {/tmp/ipykernel_68186/2067711137.py:77} Building metric engines...DONE\n",
- "2023-11-21 08:45:27,579 INFO {/tmp/ipykernel_68186/2067711137.py:81} Building simulations from 7 scenarios...\n",
- "2023-11-21 08:45:27,579 INFO {/tmp/ipykernel_68186/2067711137.py:81} Building simulations from 7 scenarios...\n",
- "2023-11-21 08:45:27,590 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:27,590 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:27,657 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "2023-11-21 08:45:27,657 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "bro im here!\n",
- "2023-11-21 08:45:28,185 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:28,185 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:28,234 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "2023-11-21 08:45:28,234 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "bro im here!\n",
- "2023-11-21 08:45:30,268 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:30,268 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:30,330 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "2023-11-21 08:45:30,330 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "bro im here!\n",
- "2023-11-21 08:45:31,045 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:31,045 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:31,107 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "2023-11-21 08:45:31,107 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "bro im here!\n",
- "2023-11-21 08:45:33,877 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:33,877 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:33,927 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "2023-11-21 08:45:33,927 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "bro im here!\n",
- "2023-11-21 08:45:34,701 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:34,701 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:34,771 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "2023-11-21 08:45:34,771 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "bro im here!\n",
- "2023-11-21 08:45:37,750 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:37,750 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:18} Building TorchModuleWrapper...\n",
- "2023-11-21 08:45:37,799 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "2023-11-21 08:45:37,799 INFO {/home/ehdykhne/nuplan-devkit/nuplan/planning/script/builders/model_builder.py:21} Building TorchModuleWrapper...DONE!\n",
- "bro im here!\n",
- "2023-11-21 08:45:38,528 INFO {/tmp/ipykernel_68186/2067711137.py:134} Building simulations...DONE!\n",
- "2023-11-21 08:45:38,528 INFO {/tmp/ipykernel_68186/2067711137.py:134} Building simulations...DONE!\n",
- "2023-11-21 08:45:38,532 INFO {/tmp/ipykernel_68186/4084133409.py:79} Running simulation...\n",
- "2023-11-21 08:45:38,532 INFO {/tmp/ipykernel_68186/4084133409.py:79} Running simulation...\n",
- "hi there\n",
- "2023-11-21 08:45:38,532 INFO {/tmp/ipykernel_68186/4231899458.py:139} Executing runners...\n",
- "2023-11-21 08:45:38,532 INFO {/tmp/ipykernel_68186/4231899458.py:139} Executing runners...\n",
- "2023-11-21 08:45:38,532 INFO {/tmp/ipykernel_68186/886081945.py:84} Starting 7 simulations using RayDistributed!\n",
- "2023-11-21 08:45:38,532 INFO {/tmp/ipykernel_68186/886081945.py:84} Starting 7 simulations using RayDistributed!\n",
- "about to start all the runners\n"
+ "2023-12-06 02:32:37,388 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/runner/executor.py:127} Number of successful simulations: 1\n",
+ "2023-12-06 02:32:37,388 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/runner/executor.py:127} Number of successful simulations: 1\n",
+ "2023-12-06 02:32:37,388 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/runner/executor.py:128} Number of failed simulations: 0\n",
+ "2023-12-06 02:32:37,388 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/runner/executor.py:128} Number of failed simulations: 0\n",
+ "2023-12-06 02:32:37,388 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/utils.py:147} Finished executing runners!\n",
+ "2023-12-06 02:32:37,388 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/utils.py:147} Finished executing runners!\n",
+ "2023-12-06 02:32:37,394 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/utils.py:74} Saved runner reports to /home/ehdykhne/Repos/Datasets/nuplan/exp/exp/simulation/closed_loop_reactive_agents/2023.12.06.02.28.50/runner_report.parquet\n",
+ "2023-12-06 02:32:37,394 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/utils.py:74} Saved runner reports to /home/ehdykhne/Repos/Datasets/nuplan/exp/exp/simulation/closed_loop_reactive_agents/2023.12.06.02.28.50/runner_report.parquet\n",
+ "2023-12-06 02:32:37,395 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/main_callback/time_callback.py:27} Simulation duration: 00:00:59 [HH:MM:SS]\n",
+ "2023-12-06 02:32:37,395 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/main_callback/time_callback.py:27} Simulation duration: 00:00:59 [HH:MM:SS]\n",
+ "2023-12-06 02:32:37,461 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/main_callback/metric_file_callback.py:79} Metric files integration: 00:00:00 [HH:MM:SS]\n",
+ "2023-12-06 02:32:37,461 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/main_callback/metric_file_callback.py:79} Metric files integration: 00:00:00 [HH:MM:SS]\n",
+ "2023-12-06 02:32:37,530 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/main_callback/metric_aggregator_callback.py:58} Running metric aggregator: closed_loop_reactive_agents_weighted_average\n",
+ "2023-12-06 02:32:37,530 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/main_callback/metric_aggregator_callback.py:58} Running metric aggregator: closed_loop_reactive_agents_weighted_average\n",
+ "2023-12-06 02:32:37,546 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/main_callback/metric_aggregator_callback.py:69} Metric aggregator: 00:00:00 [HH:MM:SS]\n",
+ "2023-12-06 02:32:37,546 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/main_callback/metric_aggregator_callback.py:69} Metric aggregator: 00:00:00 [HH:MM:SS]\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
- "\r",
- "Ray objects: 0%| | 0/7 [00:00<?, ?it/s]"
+ "Rendering histograms: 5%|â– | 1/21 [00:00<00:05, 3.84it/s]"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "\u001b[36m(wrapped_fn pid=74090)\u001b[0m about to he\n",
- "\u001b[36m(wrapped_fn pid=74090)\u001b[0m hehe\n"
+ "2023-12-06 02:32:37,714 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:37,714 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:37,717 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:37,717 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
- "Converting detections to smart agents: 0%| | 0/15 [00:00<?, ?it/s]\n",
- "Converting detections to smart agents: 0%| | 0/31 [00:00<?, ?it/s]\u001b[32m [repeated 4x across cluster] (Ray deduplicates logs by default. Set RAY_DEDUP_LOGS=0 to disable log deduplication, or see https://docs.ray.io/en/master/ray-observability/ray-logging.html#log-deduplication for more options.)\u001b[0m\n",
- "Converting detections to smart agents: 0%| | 0/37 [00:00<?, ?it/s]\u001b[32m [repeated 2x across cluster]\u001b[0m\n",
- "Converting detections to smart agents: 7%|â–‹ | 1/15 [00:18<04:20, 18.57s/it]\n",
- "Converting detections to smart agents: 13%|█▎ | 2/15 [00:20<01:55, 8.88s/it]\n",
- "Converting detections to smart agents: 20%|██ | 3/15 [00:20<00:59, 4.95s/it]\n",
- "Converting detections to smart agents: 27%|██▋ | 4/15 [00:21<00:34, 3.11s/it]\n",
- "Converting detections to smart agents: 33%|███▎ | 5/15 [00:21<00:20, 2.10s/it]\n",
- "Converting detections to smart agents: 40%|████ | 6/15 [00:21<00:13, 1.52s/it]\n",
- "Converting detections to smart agents: 53%|█████▎ | 8/15 [00:22<00:06, 1.15it/s]\n",
- "Converting detections to smart agents: 60%|██████ | 9/15 [00:22<00:04, 1.30it/s]\n",
- "Converting detections to smart agents: 87%|████████▋ | 13/15 [00:23<00:00, 2.88it/s]\n",
- "Converting detections to smart agents: 93%|█████████▎| 14/15 [00:23<00:00, 3.03it/s]\n",
- " \n",
- "Converting detections to smart agents: 5%|â– | 1/21 [00:49<16:23, 49.16s/it]\n",
- "Converting detections to smart agents: 10%|â–‰ | 2/21 [00:49<06:33, 20.70s/it]\n",
- "Converting detections to smart agents: 19%|█▉ | 4/21 [00:50<02:11, 7.76s/it]\n",
- "Converting detections to smart agents: 29%|██▊ | 6/21 [00:50<01:01, 4.12s/it]\n",
- "Converting detections to smart agents: 38%|███▊ | 8/21 [00:50<00:32, 2.48s/it]\n",
- "Converting detections to smart agents: 48%|████▊ | 10/21 [00:50<00:18, 1.66s/it]\n",
- "Converting detections to smart agents: 57%|█████▋ | 12/21 [00:51<00:10, 1.20s/it]\n",
- "Converting detections to smart agents: 67%|██████▋ | 14/21 [00:51<00:05, 1.19it/s]\n",
- "Converting detections to smart agents: 76%|███████▌ | 16/21 [00:51<00:02, 1.69it/s]\n",
- "Converting detections to smart agents: 90%|█████████ | 19/21 [00:51<00:00, 2.73it/s]\n",
- " \n",
- "Converting detections to smart agents: 26%|██▌ | 8/31 [00:51<00:38, 1.68s/it]\u001b[32m [repeated 11x across cluster]\u001b[0m\n",
- " \n",
- "Converting detections to smart agents: 61%|██████■| 19/31 [00:56<00:06, 1.94it/s]\u001b[32m [repeated 16x across cluster]\u001b[0m\n",
- "Converting detections to smart agents: 18%|█▊ | 5/28 [00:55<01:53, 4.95s/it]\n",
- "Converting detections to smart agents: 50%|█████ | 14/28 [00:56<00:08, 1.69it/s]\n",
- "Converting detections to smart agents: 94%|█████████▎| 29/31 [01:00<00:00, 2.61it/s]\n",
- "Converting detections to smart agents: 97%|█████████▋| 30/31 [01:00<00:00, 3.07it/s]\n",
- " \n",
- "Converting detections to smart agents: 51%|█████■| 20/39 [01:04<00:08, 2.37it/s]\u001b[32m [repeated 27x across cluster]\u001b[0m\n",
- "Converting detections to smart agents: 57%|█████▋ | 16/28 [01:03<00:17, 1.42s/it]\u001b[32m [repeated 18x across cluster]\u001b[0m\n",
- "Converting detections to smart agents: 93%|█████████▎| 26/28 [01:08<00:01, 1.79it/s]\n",
- "Converting detections to smart agents: 96%|█████████▋| 27/28 [01:08<00:00, 1.93it/s]\n",
- "Converting detections to smart agents: 51%|█████■| 19/37 [01:05<00:08, 2.23it/s]\u001b[32m [repeated 22x across cluster]\u001b[0m\n",
- " \n",
- " \u001b[32m [repeated 4x across cluster]\u001b[0m\n",
- "Converting detections to smart agents: 81%|████████ | 30/37 [01:11<00:03, 2.11it/s]\u001b[32m [repeated 15x across cluster]\u001b[0m\n"
+ "Rendering histograms: 62%|██████■| 13/21 [00:01<00:00, 9.57it/s]"
]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "2023-12-06 02:32:39,361 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,361 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,366 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,366 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,374 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,374 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,377 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,377 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,382 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,382 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,385 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,385 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,390 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,390 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,393 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n",
+ "2023-12-06 02:32:39,393 INFO {/opt/conda/lib/python3.9/site-packages/matplotlib/category.py:223} Using categorical units to plot a list of strings that are all parsable as floats or dates. If these strings should be plotted as numbers, cast to the appropriate data type before plotting.\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "Rendering histograms: 100%|██████████| 21/21 [00:02<00:00, 7.12it/s]\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "2023-12-06 02:32:42,303 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/main_callback/metric_summary_callback.py:344} Metric summary: 00:00:04 [HH:MM:SS]\n",
+ "2023-12-06 02:32:42,303 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/simulation/main_callback/metric_summary_callback.py:344} Metric summary: 00:00:04 [HH:MM:SS]\n",
+ "2023-12-06 02:32:42,303 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/run_simulation.py:116} Finished running simulation!\n",
+ "2023-12-06 02:32:42,303 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/run_simulation.py:116} Finished running simulation!\n"
+ ]
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 6 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 4 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 2 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 2 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 2 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 2 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 4 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 6 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 6 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 6 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 2 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 2 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 2 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 6 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 6 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 6 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 4 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 2 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk0AAAJRCAYAAABY7oO4AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAtx0lEQVR4nO3de5zWdZ3//+egHBRhPFCRgnj4aiSBCiiCqYCrqXhaN0NLF03yCEmGeGTNNkBW3VVE7NtuhZa15ffbrtWa2Te1EjaTvOEJNfHEpOIC4gwgM5w+vz+8NbdUsLf9nINwv99u1425Tp/P67rGa3zM5/rM56qpqqoKAADvqkNbDwAA8EEgmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCaA98ns2bPz6U9/uq3HAFqIaAIAKCCaYDP20EMPZeTIkRk8eHAGDhyY//t//2+SZObMmdlrr70yePDgTJ48OT169Gi+z3e+8530798/AwYMyKhRo/LSSy+96zp22223/MM//EOGDRuWXXfdNd/97ndz44035sADD8yee+6Z+++/v/m2P//5z/PJT34ygwYNypAhQ/LrX/86SbJ48eKMGDEigwYNSr9+/fLFL34xf/os8a985Sv57Gc/m+OOOy777LNPRo4cmddee22T82zYsCHjxo1L3759s++++2bQoEFpbGzMCy+8kB49emTixIkZMmRI+vXrl3vvvfcvzvan52TIkCEZOHBgDjvssDz++ONJkjVr1uScc87J3nvvnREjRuTBBx8s/M4AH0gVsFlavnx5tf/++1cvv/xyVVVVtWTJkmrXXXetfvWrX1U777xz9eqrr1ZVVVUXXnhhtdNOO1VVVVWPPfZY9ZGPfKT64x//WFVVVX3ta1+rjjnmmHddT58+faqJEydWVVVVv/vd76ptttmmuvnmm6uqqqof/OAH1dChQ6uqqqpnn322Gjp0aFVfX19VVVU988wz1c4771ytWbOmWr16dbVixYqqqqpq3bp11ahRo6o77rijqqqquuqqq6o99tijWrZsWVVVVTV69Ohq6tSpm5zn4Ycfrvr27VutX7++qqqqev3116v169dXzz//fJWkmj17dlVVVfXf//3f1Uc+8pFq5cqV7zrbAw88UB1zzDFVY2NjVVVV9etf/7oaMGBAVVVVNWPGjOqII46o1qxZU61ataoaNGhQ9Xd/93d/6VsDfEBt3dbRBrSMuXPn5rnnnsvRRx/dfFlVVbnvvvtyzDHH5MMf/nCS5Mwzz8x3v/vdJMl9992XY489NrvsskuS5Pzzz8/Xvva1VFWVmpqaTa5r9OjRSZKBAwdm9erV+cxnPpMkGTRoUJ577rkkyd13352FCxfm0EMPfct96+rq0rNnz1xyySV54IEHUlVV/ud//if77bdf8/5BRx99dHbcccckydChQ/PYY49tcpY99tgja9euzec///mMGDEio0aNSocOb25U79SpU04//fQkyUEHHZSePXvmkUceyfz58zc525133plHHnkkQ4YMab58yZIlWbNmTe67776MGTMmHTt2TMeOHXPaaaflgQce2ORswAebaILNVFVVGTBgwFveZkqSG2+8cZNvub09jt4tlP5cly5dkiRbbbXVO86vW7euedlHHXVUbrvttnfc/2tf+1qWLVuWBx98MF26dMlFF12UxsbGdyz/7cvcmNra2jzxxBP51a9+lfvuuy+XXXZZfv3rX2frrTf+466mpuZdZ6uqKp///Ofz1a9+daPXAVsO+zTBZmrYsGF55pln3rLfzvz58zN8+PDcddddWbp0aZLk1ltvbb7+8MMPz1133ZXFixcnSb7+9a/n8MMPL46nd3PkkUfm7rvvbt4fKEl+97vfJUmWL1+enj17pkuXLnn11Vdzxx13/NXrWbJkSVatWpUjjzwyU6dOzW677ZYFCxYkeXMfpNtvv7153YsXL86AAQPedbbjjjsut912W+rq6pK8uc/UvHnzkrz5fH3nO9/JunXrsnr16nzve9/7q+cG2j9bmmAztcMOO+QnP/lJLr744nzpS1/K2rVrs+uuu+Y///M/M2nSpBx00EH56Ec/mpEjR6a2tjZJ0q9fv0ybNi1HHnlkkqR37975xje+8b7Ms9dee+W73/1uxo4dm9WrV2fNmjUZOHBgbr/99nzxi1/MySefnP322y+77LJL/uZv/uavXk9dXV2+8IUvZO3atdmwYUOGDRuWo48+Oi+99FJ22mmnLFy4MEOGDMnKlSvzve99L127dn3X2Q499NBMnTo1J5xwQtavX5+1a9dm1KhRGTx4cM4+++w8+uij2WeffdKrV68ccsghefHFF9+X5wtof2oq25dhi7NixYp069YtyZt/nbZw4cLm/Zo2Vy+88EIGDx7cvIUN4L2ypQm2QJdeemnmzJmTNWvWZPfdd8+//uu/tvVIAO2eLU3AX/Rv//ZvmTlz5jsuv+mmm3LIIYe0wUTtcyZg8yaaAAAK+Os5AIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJngXjY2NGT58+P+vZZxxxhl5/PHH35+BAGgzogkAoIBoIklSVVXGjx+fESNG5Igjjsgf//jHTJ06NUOHDs348eNzwAEHJEkeffTRHHzwwRk2bFi+9rWvbXRZs2fPzoknnphRo0bl4IMPzne+852MGjUqhx12WJqamja6riT53Oc+l+HDh+eTn/xkFi1alCQZOHBgzj333AwZMiTTpk3b6Pr+8z//MwceeGCGDx+eW265JUmyzz77ZMyYMRk8eHC++93vJkmee+65fOpTn8rw4cPzpS99KcmbW5JOO+20jBw5Mscff3waGhqSJBdccEEOO+ywXHXVVRtd55o1a3LwwQc3nz/11FPz7LPP5tprr82IESMyaNCg/OIXv3jLfe6///5MnDgxSfLUU0/ljDPOSJLcfffdOeSQQzJs2LB8//vff5fvEgBtqoKqqn7yk59UkydPrqqqqubNm1edddZZ1aGHHlpt2LChevrpp6vdd9+9qqqqGjVqVLVgwYJqw4YN1RFHHFE9//zz71jWt7/97Wrs2LFVVVXVlVdeWU2YMKGqqqqaMGFC9fOf//wd67rggguqqqqqVatWVVVVVXfeeWd1+eWXV1VVVbvvvnv1wgsvVOvXr68GDBiw0dlPO+206rHHHquqqqrWr19fVVVVde3atVq2bFnV2NhYDRw4sFq3bl316U9/ulq4cGFVVVU1bty46qGHHqpuuumm6pvf/GZVVVX1f/7P/6muvfba6qGHHqpOPfXUqqqq6p577qkOO+ywja7385//fPXoo49Wb7zxRjV8+PC3PIYlS5Y0XzZmzJjqscceq+67777qy1/+clVVVfXkk09WY8aMqdavX18NHTq0ampqqtavX18deuih1bp16zb1bQKgDW3d1tFG+7BgwYL8x3/8R37961+nqqp07NgxAwYMSE1NTfbee+/U1tYmSV599dV8/OMfT/LmVqBnn302u+222zuWN2DAgCTJLrvsks6dOzd/vXz58rz44otvWVfv3r2zfv36XHLJJZk/f36amprSr1+/JMkOO+yQPn36JEm22Wabjc4+efLk/Mu//EtWrVqV888/PwcddFB233337LjjjkmSXXfdNUuXLs3TTz+ds846K0myYsWKHH744VmwYEEeeuih3HbbbVm7dm0OOeSQLFy4MIMGDUqSHHjggZt8zkaPHp0f/OAH2XfffXPssccmSW6//fbcdttt6dChQxYvXvyW29fU1DR/XVVVkmTp0qV55plncuSRRzafX7JkSXr27LnJ9QLQNkQTSZK+ffvmM5/5TCZPnpwkqaury+mnn56qqvLss8+mvr4+SfKRj3wkTz75ZPr27ZuHH34455577kaX9+eB8PZYePu61q5dm/nz5+fVV1/Nb37zm/z4xz/Oj370o3fcd1N69+6dW265JS+99FJOP/303HvvvXnhhReyfPnydO3aNXV1denRo0c+9rGP5brrrkufPn1SVVXWr1+fRYsWZejQoTn99NObZ3nkkUfy05/+NEkyb968Ta535MiRufrqq/OHP/wh119/fZLkuuuuyxNPPJHly5fnk5/85Ftuv8MOO6Suri5J8vvf/z5J0qNHj3z84x/PL37xi3Ts2DFr165Nx44d/+JjBqD1iSaSJMcdd1zuvffejBgxIkly2mmn5fDDD8+wYcOy//77p0ePHkmSKVOmZOzYsamqKqNGjdroVqa/Zl2nnHJKXnnllRxxxBHNW7JKXX311fnv//7vrFixonmfod69e+eLX/xinnzyyUyYMCFbbbVVpk+fnnPPPTdNTU3p0KFDvvWtb+Xss8/O2WefnW9/+9tJki9/+csZNWpUunfvnkMPPTRDhgzZ5Hq33nrr9O/fP08//XR69+6dJBkxYkQOOeSQHHjggenevftbbt+/f/80Njbm8MMPz1577ZUk6dChQ6644or8zd/8TTp06JAPfehD+eEPf/ieHj8AraOm+tP7BPA2f9rq8cwzz2TChAn5r//6r7YeqdjgwYPfdSsRALxXtjSxSVdddVXmzJmT1atXZ9asWRu9za9+9at3/IXZ/fff32Iztfb6/uTGG2/Mf/zHfzSf32+//XLDDTe0+HoBaD9saQIAKOA4TQAABUQTAEAB0QQAUEA0AQAU2Kz+em7Dhg15+eWX061bt6KDIgKbj6qqsmLFiuy8887p0MHvg8D7b7OKppdffrn5IIPAlqmuri69evVq6zGAzdBmFU3dunVL8uYPzbcfjRnYvDU0NKR3797NPwcA3m+bVTT96S257t27iyY2Kxs2bMiaNWvaeox2qVOnTm95O85b80BL2ayiCTZHGzZsyIsvvpjGxsa2HqVd6tKlS/r06dPWYwBbANEE7dyaNWvS2NiYnj17ZptttmnrcdqV1atXZ/HixbbCAa1CNMEHxDbbbJNle+5VfPtdXqp7x2Vf+cpX0r1799xzzz3p1q1bhg8fnhUrVuTSSy/Nvvvumx/96EdJkpNOOil33nln7r777px77rm59NJLc80117xlWRu7LEl+9KMf5aSTTnrLbY488sgMHTo0++23Xx588MGN3g+gvRNNsIU5//zzs3LlynTo0CEXXHBBbrvttjzxxBMZOXJk7rzzzlRVlcMPPzxJcu+996axsTHz5s3LCy+8kF/84hc54ogjcs8992TRokX5t3/7tyxatCgjR47MHXfckTPPPDO/+93v0tDQkFWrVmXOnDlJkoEDB+bKK6/MpZdemo4dO2b9+vW5+eab89RTT+WKK67IlClTMnTo0DzxxBM599xzc8UVV+R//a//lWOPPTYPP/xw6uvrU1VVhgwZ0ryewYMHt+XTCGyBHMwEtjDf+MY3stNOO6VHjx5J3jy+UZJ07tw5a9euzfr169OpU6ckyciRIzNhwoQMHjw4u+22W+rq6nL77bfns5/9bLp165axY8emvr4+SfKpT32qOWQeeeSRXHDBBfn4xz+eJJk/f36uv/76fOELX0jy5ttqGzZsyI477pjf//736dq1a04//fTmGYcNG5YzzjgjDz/8cO677758+MMfzqpVq96xHoDWZEsTbGHOPvvsLF68OJMmTUpjY2NWr16dfv36JUnGjh2bJLn22ms3et9DDjkkv/zlL7PddttlxYoV+frXv57tttsuSZpDK0n23Xff3H777Xn66aeTJPvtt18uvfTS5uv/53/+J6+//nrWr1+fDRs2ZKuttnrLerbaaqvU1NSkqqoMHz48y5Ytaw6wP18PQGuqqf70a+ZmoKGhIbW1tamvr3fIATYbjY2Nef7557P77runS5cubTbHmjVrctVVV+ULX/hC9thjjzab48/9+XOzZs0ar3+gRdnSBBTp1KlTpk2b1tZjALQZ0QQfEKtXr27rEdodzwnQmkQTtHOdOnVKly5dsnjx4rYepV3q0qVLOnXq5FhNQIsTTdDOdejQIX369BEFm/D2j1EBaCmiCT4AOnTo0KY7gQPgOE0AAEVEEwBAAdEEAFBANAEAFBBNAAAFRBMAQAHRBABQoEWj6bnnnstZZ52VU045JUlyww03ZNy4cTnvvPNSVVUWLFiQ008/PZ/73OeyYMGCzJ07NxMmTGj+fKsrr7wyy5Yta8kRAQCKtGg07bHHHvnmN7+Z5M1PSJ8/f35mzpyZfv36Zc6cObnxxhsza9aszJo1KzfeeGMeeOCBXHXVVamvr8+cOXPSt2/f7LTTTptcflNTUxoaGt5yAgBoCa12RPBly5alR48eSZI+ffqkrq4uK1asSLdu3ZIkK1asyKmnnpobbrghAwcOzA9/+MPsu++++cpXvpKJEydmu+22e8cyp02blquvvrq1HgIfAC/t0vs93X6Xl+paaBIANjettk/TTjvtlKVLlyZJFi1alF69eqVbt25ZsWJFGhoa0q1bt/Tu3TtXX3116urqMm7cuDz99NM57rjj8rOf/Wyjy7zssstSX1/ffKqr8z9AAKBltOiWpmXLluWKK67IvHnzcsMNN2TAgAGZMGFCGhsbc/7552eHHXbIuHHjUlVVJk2alCRZuHBhkmSvvfbKhg0bcuutt+bCCy/c6PI7d+6czp07t+RDAABIktRUVVW19RDvl4aGhtTW1qa+vj7du3dv63FoA96e23J5/QMtzSEHAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACmzdmitbtGhRxo8fn5122ikf+9jH0rlz5yxcuDDr16/PrFmzMnXq1CxZsiSnnHJK+vbtm+nTp2fatGmtOSIAwEa16pamp59+OieccEK+9a1v5ZFHHsn8+fMzc+bM9OvXL3PmzMnKlSszefLk/OY3v8l1112XiRMnvuvympqa0tDQ8JYTAEBLaNVoGjhwYL7//e/nmGOOSb9+/dKjR48kSZ8+fVJXV5cDDjggM2bMyK677pqdd945N910U2bPnr3J5U2bNi21tbXNp969e7fSIwEAtjStGk3f/va389WvfjV33XVX5s2bl6VLlyZ58227Xr165aSTTsqVV16ZuXPnpkePHjn++OPz1FNPbXJ5l112Werr65tPdXV1rfVQAIAtTKvu03TUUUflq1/9am699dbsueee2XnnnTNhwoQ0Njbm/PPPT5LMmDEj48ePT1VVmTFjRrbddttNLq9z587p3Llza40PAGzBaqqqqtp6iPdLQ0NDamtrU19fn+7du7f1OLSBl3Z5b2/R7vKSrZObC69/oKU55AAAQAHRBABQQDQBABQQTQAABUQTAEAB0QQAUEA0AQAUEE0AAAVEEwBAAdEEAFBANAEAFBBNAAAFRBMAQAHRBABQQDQBABQQTQAABUQTAEAB0QQAUEA0AQAUEE0AAAVEEwBAAdEEAFBANAEAFBBNAAAFRBMAQAHRBABQQDQBABQQTQAABUQTAEAB0QQAUEA0AQAUEE0AAAVEEwBAAdEEAFBANAEAFBBNAAAFRBMAQAHRBABQQDQBABQQTQAABUQTAEAB0QQAUEA0AQAUEE0AAAVEEwBAAdEEAFBANAEAFBBNAAAFRBMAQAHRBABQQDQBABQQTQAABUQTAEAB0QQAUEA0AQAUEE0AAAVEEwBAAdEEAFBANAEAFBBNAAAFRBMAQAHRBABQQDQBABQQTQAABUQTAEAB0QQAUEA0AQAUEE0AAAVEEwBAAdEEAFBANAEAFBBNAAAFRBMAQAHRBABQQDQBABQQTQAABUQTAEAB0QQAUEA0AQAUEE0AAAVEEwBAAdEEAFBANAEAFBBNAAAFRBMAQAHRBABQQDQBABQQTQAABUQTAECBrVtzZRs2bMjkyZNTX1+fQYMGpb6+PgsXLsz69esza9asTJ06NUuWLMkpp5ySvn37Zvr06Zk2bVprjggAsFGtuqXpzjvvzEsvvZSqqrLLLrtk/vz5mTlzZvr165c5c+Zk5cqVmTx5cn7zm9/kuuuuy8SJE1tzPACATWrVLU1PP/10DjrooJxzzjkZOXJkBg0alCTp06dP6urqcsABB2TGjBnZZ599smzZstx0003ZbbfdcsYZZ2x0eU1NTWlqamo+39DQ0BoPAwDYArXqlqZevXplxx13TE1NTXbYYYcsXbo0SbJo0aL06tUrJ510Uq688srMnTs3PXr0yPHHH5+nnnpqk8ubNm1aamtrm0+9e/durYcCAGxhaqqqqlprZW+88UbGjx+fbbfdNnvvvXfWrl2bRYsWpbGxMbfccktqampy/fXX54QTTkhVVZkxY0a23XbbTJ8+faPL29iWpt69e6e+vj7du3dvrYdFO/LSLu8tnHd5qa6FJqG1NTQ0pLa21usfaDGtGk0tzQ9NRNOWy+sfaGkOOQAAUEA0AQAUEE0AAAVEEwBAAdEEAFBANAEAFBBNAAAFRBMAQAHRBABQoPgDe5csWZLf//73Wb58efbee+/mD9sFANgSFEXTpEmT0rlz5+yzzz7Zfvvt8+CDD2b27Nn59Kc/ncMOO6ylZwQAaHNF0TRlypR07NjxHZevXbv2fR8IAKA9KoqmPwXTT37yk8ydOzfbb799Lrnkko2GFADA5qhoR/Drr78+69aty0MPPZRJkyZl6dKlLT0XAEC7UhRNJ510UiZNmpS+ffvmm9/8Zk488cQWHgsAoH0piqaampoMGzYsK1euzOuvv55+/fq19FwAAO1KUTRdd9112XPPPfPqq69m4sSJufbaa1t6LgCAdqUomrbZZpvccccd6dWrV7bffvtMmTKlpecCAGhXiv56zpYlAGBLVxRN5557bvr27Zv9998/tbW1ee6553L//ffn2GOPzZFHHtnSMwIAtLmiaPr617+exx57LA899FCWL1+ej33sY5k2bVq6du3a0vMBALQLxZ89179///Tv378lZwEAaLeKdgQHANjSFUfTI4880pJzAAC0a8XRdM899+Tv//7v84Mf/CDr169vyZkAANqdmqqqqtIbv/766zn11FOzatWqnHbaaTn77LNbcrb3rKGhIbW1tamvr0/37t3behzawEu79H5Pt9/lpboWmoTW5vUPtLTiHcEvuuiiNDU1ZcqUKRk4cGAmTZrUknMBALQrxVuaXnnllXz0ox9N8uYWp+23374l5/qr+E0TW5q2XF7/QEsr3qfpuuuua/562rRpLTIMAEB7VRxNr7/++ka/BgDYEhTv03TyySfnM5/5TDp06JAzzzyzJWcCAGh3iqPpqKOOSt++fdPU1JSampqWnAkAoN0pjqa///u/T58+fbL11lunpqYm//AP/9CScwEAtCvF0bT//vvnS1/6UkvOAgDQbhVH06233pr7778/Xbt2TZJ873vfa7GhAADam+Jomj9/fguOAQDQvhUfcuCaa67JmDFjkiSXX355iw0EANAeFUfTK6+8kj333DNJsm7duhYbCACgPSqOpg4dOuTVV1/N3XffncWLF7fkTAAA7U5xNF199dXp379/nn/++dx8880tORMAQLtTHE3Tpk3Liy++mEWLFvnsOQBgi1P813PnnntukuSNN97I7bff3mIDAQC0R8XR9NGPfjRJsnbt2qxcubLFBgIAaI+Ko+mcc85JTU1NOnXqlBNPPLEFRwIAaH+Ko+nSSy9t/rqmpiZ/+MMfsvfee7fIUAAA7U1xNJ133nnp169fkuSJJ57I8OHDfWgvALDFKI6mAw88MNdcc02S5IorrhBMAMAWpTiaXn/99UyZMiU1NTVZtmxZS84EANDuFEfT17/+9Tz++OOpqir9+/dvyZkAANqd9/SBvddee2369+/vA3sBgC2OD+wFACjgA3sBAAoUR9OZZ56Z/v3757nnnsvMmTNbciYAgHaneEfwn//857n44otbchYAgHarOJq+853v5P77709tbW2S5Hvf+16LDQUA0N4URdNPf/rTPProo3n88cfziU98oqVnAgBod4r2abrrrruSJLNmzWrRYQAA2quiaFq0aFHuueee5n/vueeelp4LAKBdKXp77uSTT84rr7zS/G9NTU1LzwUA0K4URdOYMWNaeg4AgHat+DhNAABbMtEEAFBANAEAFBBNAAAFRBMAQAHRBABQQDQBABQQTQAABUQTAEAB0QQAUEA0AQAUEE0AAAVEEwBAAdEEAFBANAEAFBBNAAAFRBMAQAHRBABQQDQBABQQTQAABUQTAEAB0QQAUEA0AQAUEE0AAAVEEwBAAdEEAFBANAEAFBBNAAAFRBMAQAHRBABQoNWjadWqVRk0aFDuvvvu3HDDDRk3blzOO++8VFWVKVOmZMKECfntb3+b119/PZdddllrjwcAsFGtHk3Tp0/P6NGjs2bNmsyfPz8zZ85Mv379MmfOnKxcuTKTJ0/Ob37zm1x33XWZOHHiuy6rqakpDQ0NbzkBALSEVo2me+65J5/4xCfy4Q9/OKtWrUqPHj2SJH369EldXV0OOOCAzJgxI7vuumt23nnn3HTTTZk9e/Ymlzdt2rTU1tY2n3r37t1KjwQA2NJs3Zoru/fee/P666/n6aefTpcuXfKRj3wkSbJo0aIMGDAghxxySI477rhMnDgxBx98cA466KD88Ic/3OTyLrvsslx00UXN5xsaGoQTANAiWjWarrnmmiTJ7Nmz07NnzyxYsCATJkxIY2Njzj///CTJjBkzMn78+FRVlRkzZmTbbbfd5PI6d+6czp07t8rsAMCWraaqqqqth3i/NDQ0pLa2NvX19enevXtbj0MbeGmX97alcZeX6lpoElqb1z/Q0hxyAACggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoMDWrbmyH//4x/nJT36SJUuWZPz48XnssceycOHCrF+/PrNmzcrUqVOzZMmSnHLKKenbt2+mT5+eadOmteaIAAAb1arRdPzxx+f444/P8uXLc9FFF6WqqsyePTszZ87MnDlzsnLlykyePDnf+ta38tOf/jQTJ05szfEAADapTd6emzJlSsaOHZsePXokSfr06ZO6uroccMABmTFjRnbdddfsvPPOuemmmzJ79uxNLqepqSkNDQ1vOQEAtIRWj6bLL788Rx99dA444IAsXbo0SbJo0aL06tUrJ510Uq688srMnTs3PXr0yPHHH5+nnnpqk8uaNm1aamtrm0+9e/durYcBAGxhaqqqqlprZbNmzcq//uu/ZsiQIdlvv/3yxhtvZNGiRWlsbMwtt9ySmpqaXH/99TnhhBNSVVVmzJiRbbfdNtOnT9/o8pqamtLU1NR8vqGhIb179059fX26d+/eWg+LduSlXd5bOO/yUl0LTUJra2hoSG1trdc/0GJaNZpamh+aiKYtl9c/0NIccgAAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKCAaAIAKCCaAAAKiCYAgAKiCQCggGgCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKDA1m258lWrVmX8+PHZeuutM2LEiNx3333Zeuutc9NNN+WOO+5Iz549M3z48LYcEQAgSRtH049+9KOcfPLJOfroozNy5MiMHz8+zz33XJ599tk8+uijOeWUU9pyPACAZm0aTX/84x+z//77J0l69uyZBQsWpGvXrrn11lszePDgTJ48Oaecckr69eu30fs3NTWlqamp+Xx9fX2SpKGhoeWHp11asWHDe7q9/1Y2H3/6XlZV1caTAJurNo2mXr165Y9//GM+8YlPZMOGDbniiivywAMP5MUXX8yDDz6Ya665JpdffnmmTZu20ftPmzYtV1999Tsu7927d0uPzuaitratJ+B9tmLFitT6vgItoKZqw1/L/rRPU+fOnfPJT34yJ598ci6++OLceOONueWWW/Lyyy9n3333zac//emN3v/tW5o2bNiQ1157LTvttFNqampa62G8Q0NDQ3r37p26urp07969zeb4a5i99X1Q507a1+xVVWXFihXZeeed06GDv3EB3n9tGk2bq4aGhtTW1qa+vr7N/0fyXpm99X1Q504+2LMDvFd+HQMAKCCaAAAKiKYW0Llz51x11VXp3LlzW4/ynpm99X1Q504+2LMDvFf2aQIAKGBLEwBAAdEEAFBANAEAFBBN75PnnnsuZ5111js+L+9b3/pWTjvttJx00kl55JFH2mi6Tfvxj3+cL3zhCznxxBPzy1/+svny+++/P2eccUZOO+20vPLKK2044aZtavZLLrkkY8eOzcknn5yVK1e24YQbt6m5k+TnP/95+vbt20aT/WWbmv3JJ5/MuHHjMm7cuCxYsKANJwRoQRXvq9GjR7/l/BlnnFGtXr26mjt3bjVr1qw2muove+2116qzzz67+fxnP/vZasOGDdVjjz1W/eM//mMbTvaXvX32P/nnf/7nau7cuW0wUZm3z/3aa69Vl1122Tv+G2qP3j772LFjq4svvrgaN25ctXTp0jacDKDl2NLUwv7u7/4uxx57bCZNmpTjjjuurcfZpClTpuScc85pPl9VVWpqatKnT5/U1dW14WR/2dtnT5JXX301Dz/8cA488MA2muove/vcU6dOzcUXX9yGE5V7++wPPfRQLr/88pxzzjm54YYb2m4wgBYkmlrYt7/97fy///f/8u///u+57rrr2nqcjbr88stz9NFHZ+DAgc2X1dTUpKqqLFq0KL169WrD6d7dxmZ/5ZVXcvHFF2fmzJnZaqut2nC6TXv73KtWrcqCBQty2WWXZd68ebn11lvbeMJN29hzvscee2S77bbLjjvu2C7fEgV4P2zd1gNsLpYtW5Yrrrgi8+bNyz/90z9lwYIFmT17dg4++OCcffbZqa+vz9ixY9t6zHeYNWtWfvazn+W1117LM888k9/+9reZPXt2zjrrrIwdOzZr1qzJ9OnT23rMjdrU7CeeeGJ23333XHLJJTnvvPOy7777tvWob7Gpuf/rv/4rSXLKKadkzJgxbTzlxm1q9gkTJuScc85JU1NTrrzyyrYeE6BFOLglAEABb88BABQQTQAABUQTAEAB0QQAUEA0AQAUEE0AAAVEE61u9uzZufvuu9/1Nu/nkTBa86gaZ5xxRhobG1ttfQC0Hge35H0ze/bs3Hfffdlmm23y0Y9+NOvWrctjjz2W888/P3fddVfeeOONjBkzJnPnzs0bb7yRJNl6663z05/+tPm6Z555Jvfff38GDRqUNWvW5Pnnn09tbW2mTJmy0XXtvffe2W677XLhhRdmypQpWbJkSRobG3PjjTfm5JNPztChQzNkyJCMHDmy+b7f//7388tf/jLbbbddpk+fnjvvvDNz585NQ0NDLr300jQ0NGTGjBnp0aNHzjvvvNx777159NFH09DQkBtuuCE33XRT3njjjTQ1NWXkyJE56qijcuaZZ6ZPnz55/PHHW/U5B6D1iCbeV8ccc0xGjx6dww8/PL/85S8zffr03Hzzzdl///3TvXv3PPjggxk2bFh69uyZo446KieccMJbrttxxx1z9NFHZ/To0fnyl7+cAw44IMccc8xG13XUUUfl1FNPzec+97k8+eSTeeCBBzJkyJC8/PLLefbZZ1NVVSZNmvSOj1Kpq6vLgAED8rd/+7fp3Llzbr755owYMSKdOnXKvHnz8rOf/Szf+MY3ss022yR582ND7rjjjvzqV7/K97///STJ6NGjM2DAgIwdOzYdOnTIpz71qYwZMybz589v0ecXgLYjmnhf1dbWJkk+9KEPJUk6d+6c1atXZ/Lkyc3xcttttzXfvqqqt1w3e/bs5mVce+21efDBB3P66afnjjvuSLdu3d6yrrVr1zb/u2HDhvTv3z9f+cpXmq/v2rXrRj97btKkSZk/f36+9KUv5ZprrknXrl3fcr+f/exnqampecf9/vyyrl27pmPHjmlqakpVVencuXPz4wVg8ySaaHEHH3xwxo4dm+233z6HHXZY9t1330ybNi3r1q3LBRdc8Jbr/tz06dOzdOnSfOhDH2re6vPn7rnnnsyfPz8HHnhg+vXrlw0bNuSiiy7KG2+8kalTp25ynv/9v/93/vCHP6Rjx47ZcccdM3r06Jxzzjnp0qVLTj755Fx44YU5//zz06NHj5x99tkZMWJEvvjFL2b58uX5l3/5l8ycOfMtyzvyyCNz4YUXZvHixXnhhRfel+cMgPbHZ8/xgTR79uzmt/gAoDXY0sQHwt13353f/va3zecvv/zydOrUqcXvCwB/YksTAEABx2kCACggmgAACogmAIACogkAoIBoAgAoIJoAAAqIJgCAAqIJAKDA/weM8gATux5rPAAAAABJRU5ErkJggg==\n",
+ "text/plain": [
+ "<Figure size 600x600 with 2 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "image/png": "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\n",
+ "text/plain": [
+ "<Figure size 600x600 with 6 Axes>"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
}
],
"source": [
- "# Run the simulation loop (real-time visualization not yet supported, see next section for visualization)\n",
- "main(cfg)\n",
+ "from nuplan.planning.script.run_simulation import main as main_simulation\n",
"\n",
- "# Simple simulation folder for visualization in nuBoard\n",
- "simple_simulation_folder = cfg.output_dir"
+ "# Run the simulation loop (real-time visualization not yet supported, see next section for visualization)\n",
+ "main_simulation(cfg)"
]
},
{
- "cell_type": "code",
- "execution_count": null,
- "id": "c221b637",
+ "cell_type": "markdown",
+ "id": "7faa00c1",
"metadata": {},
+ "source": [
+ "## Prepare the nuBoard config"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 19,
+ "id": "1bdb7bef",
+ "metadata": {
+ "scrolled": true
+ },
"outputs": [],
"source": [
- "from nuplan.planning.script.run_simulation import main as main_simulation\n",
+ "# Location of path with all nuBoard configs\n",
+ "CONFIG_PATH = '../nuplan/planning/script/config/nuboard'\n",
+ "CONFIG_NAME = 'default_nuboard'\n",
"\n",
- "# Run the simulation loop (real-time visualization not yet supported, see next section for visualization)\n",
- "main_simulation(cfg)"
+ "# Initialize configuration management system\n",
+ "hydra.core.global_hydra.GlobalHydra.instance().clear() # reinitialize hydra if already initialized\n",
+ "hydra.initialize(config_path=CONFIG_PATH)\n",
+ "\n",
+ "# Compose the configuration\n",
+ "cfg = hydra.compose(config_name=CONFIG_NAME, overrides=[\n",
+ " 'scenario_builder=nuplan_mini', # set the database (same as simulation) used to fetch data for visualization\n",
+ " f'simulation_path={[output_folder]}', # nuboard file path(s), if left empty the user can open the file inside nuBoard\n",
+ "])\n"
]
},
{
- "cell_type": "code",
- "execution_count": null,
- "id": "f87e05c8",
+ "cell_type": "markdown",
+ "id": "99c14523",
"metadata": {},
- "outputs": [],
- "source": []
+ "source": [
+ "## Launch nuBoard (open in new tab - recommended)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 20,
+ "id": "6c82ee09",
+ "metadata": {
+ "scrolled": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "2023-12-06 02:41:59,781 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:18} Building AbstractScenarioBuilder...\n",
+ "2023-12-06 02:41:59,781 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:18} Building AbstractScenarioBuilder...\n",
+ "2023-12-06 02:41:59,792 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:21} Building AbstractScenarioBuilder...DONE!\n",
+ "2023-12-06 02:41:59,792 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:21} Building AbstractScenarioBuilder...DONE!\n",
+ "2023-12-06 02:41:59,793 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/nuboard.py:84} Opening Bokeh application on http://localhost:5006/\n",
+ "2023-12-06 02:41:59,793 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/nuboard.py:84} Opening Bokeh application on http://localhost:5006/\n",
+ "2023-12-06 02:41:59,793 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/nuboard.py:85} Async rendering is set to: True\n",
+ "2023-12-06 02:41:59,793 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/nuboard.py:85} Async rendering is set to: True\n",
+ "2023-12-06 02:41:59,794 INFO {/opt/conda/lib/python3.9/site-packages/bokeh/server/server.py:403} Starting Bokeh server version 2.4.3 (running on Tornado 6.2)\n",
+ "2023-12-06 02:41:59,794 INFO {/opt/conda/lib/python3.9/site-packages/bokeh/server/server.py:403} Starting Bokeh server version 2.4.3 (running on Tornado 6.2)\n",
+ "2023-12-06 02:41:59,794 WARNING {/opt/conda/lib/python3.9/site-packages/bokeh/server/util.py:145} Host wildcard '*' will allow connections originating from multiple (or possibly all) hostnames or IPs. Use non-wildcard values to restrict access explicitly\n",
+ "2023-12-06 02:41:59,794 WARNING {/opt/conda/lib/python3.9/site-packages/bokeh/server/util.py:145} Host wildcard '*' will allow connections originating from multiple (or possibly all) hostnames or IPs. Use non-wildcard values to restrict access explicitly\n",
+ "2023-12-06 02:41:59,794 INFO {/opt/conda/lib/python3.9/site-packages/bokeh/server/tornado.py:360} User authentication hooks NOT provided (default user enabled)\n",
+ "2023-12-06 02:41:59,794 INFO {/opt/conda/lib/python3.9/site-packages/bokeh/server/tornado.py:360} User authentication hooks NOT provided (default user enabled)\n",
+ "2023-12-06 02:42:05,418 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/simulation_tile.py:172} Minimum frame time=0.017 s\n",
+ "2023-12-06 02:42:05,418 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/simulation_tile.py:172} Minimum frame time=0.017 s\n",
+ "2023-12-06 02:42:05,423 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/tabs/scenario_tab.py:485} Rending scenario plot takes 0.0006 seconds.\n",
+ "2023-12-06 02:42:05,423 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/tabs/scenario_tab.py:485} Rending scenario plot takes 0.0006 seconds.\n",
+ "2023-12-06 02:42:05,531 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET / (10.40.117.44) 267.62ms\n",
+ "2023-12-06 02:42:05,531 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET / (10.40.117.44) 267.62ms\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "INFO:tornado.access:200 GET / (10.40.117.44) 267.62ms\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "2023-12-06 02:42:06,039 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 101 GET /ws (10.40.117.44) 1.62ms\n",
+ "2023-12-06 02:42:06,039 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 101 GET /ws (10.40.117.44) 1.62ms\n",
+ "2023-12-06 02:42:06,041 INFO {/opt/conda/lib/python3.9/site-packages/bokeh/server/views/ws.py:132} WebSocket connection opened\n",
+ "2023-12-06 02:42:06,041 INFO {/opt/conda/lib/python3.9/site-packages/bokeh/server/views/ws.py:132} WebSocket connection opened\n",
+ "2023-12-06 02:42:06,042 INFO {/opt/conda/lib/python3.9/site-packages/bokeh/server/views/ws.py:213} ServerConnection created\n",
+ "2023-12-06 02:42:06,042 INFO {/opt/conda/lib/python3.9/site-packages/bokeh/server/views/ws.py:213} ServerConnection created\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "INFO:tornado.access:101 GET /ws (10.40.117.44) 1.62ms\n"
+ ]
+ },
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "2023-12-06 02:42:06,261 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/simulation_tile.py:172} Minimum frame time=0.017 s\n",
+ "2023-12-06 02:42:06,261 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/simulation_tile.py:172} Minimum frame time=0.017 s\n",
+ "2023-12-06 02:42:06,267 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/tabs/scenario_tab.py:485} Rending scenario plot takes 0.0007 seconds.\n",
+ "2023-12-06 02:42:06,267 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/tabs/scenario_tab.py:485} Rending scenario plot takes 0.0007 seconds.\n",
+ "2023-12-06 02:42:06,370 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET / (10.40.117.44) 261.77ms\n",
+ "2023-12-06 02:42:06,370 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET / (10.40.117.44) 261.77ms\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "INFO:tornado.access:200 GET / (10.40.117.44) 261.77ms\n"
+ ]
+ },
+ {
+ "ename": "KeyboardInterrupt",
+ "evalue": "",
+ "output_type": "error",
+ "traceback": [
+ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+ "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)",
+ "Cell \u001b[0;32mIn[20], line 4\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mnuplan\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mplanning\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mscript\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mrun_nuboard\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m main \u001b[38;5;28;01mas\u001b[39;00m main_nuboard\n\u001b[1;32m 3\u001b[0m \u001b[38;5;66;03m# Run nuBoard\u001b[39;00m\n\u001b[0;32m----> 4\u001b[0m \u001b[43mmain_nuboard\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcfg\u001b[49m\u001b[43m)\u001b[49m\n",
+ "File \u001b[0;32m/opt/conda/lib/python3.9/site-packages/hydra/main.py:44\u001b[0m, in \u001b[0;36mmain.<locals>.main_decorator.<locals>.decorated_main\u001b[0;34m(cfg_passthrough)\u001b[0m\n\u001b[1;32m 41\u001b[0m \u001b[38;5;129m@functools\u001b[39m\u001b[38;5;241m.\u001b[39mwraps(task_function)\n\u001b[1;32m 42\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mdecorated_main\u001b[39m(cfg_passthrough: Optional[DictConfig] \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m Any:\n\u001b[1;32m 43\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m cfg_passthrough \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m---> 44\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mtask_function\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcfg_passthrough\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 45\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m 46\u001b[0m args \u001b[38;5;241m=\u001b[39m get_args_parser()\n",
+ "File \u001b[0;32m~/Repos/nuplan-devkit/nuplan/planning/script/run_nuboard.py:74\u001b[0m, in \u001b[0;36mmain\u001b[0;34m(cfg)\u001b[0m\n\u001b[1;32m 69\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m 70\u001b[0m \u001b[38;5;124;03mExecute all available challenges simultaneously on the same scenario.\u001b[39;00m\n\u001b[1;32m 71\u001b[0m \u001b[38;5;124;03m:param cfg: DictConfig. Configuration that is used to run the experiment.\u001b[39;00m\n\u001b[1;32m 72\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m 73\u001b[0m nuboard \u001b[38;5;241m=\u001b[39m initialize_nuboard(cfg)\n\u001b[0;32m---> 74\u001b[0m \u001b[43mnuboard\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n",
+ "File \u001b[0;32m~/Repos/nuplan-devkit/nuplan/planning/nuboard/nuboard.py:112\u001b[0m, in \u001b[0;36mNuBoard.run\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 110\u001b[0m \u001b[38;5;66;03m# Catch RuntimeError in jupyter notebook\u001b[39;00m\n\u001b[1;32m 111\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 112\u001b[0m \u001b[43mio_loop\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mstart\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 113\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mRuntimeError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m 114\u001b[0m logger\u001b[38;5;241m.\u001b[39mwarning(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00me\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n",
+ "File \u001b[0;32m/opt/conda/lib/python3.9/site-packages/tornado/platform/asyncio.py:215\u001b[0m, in \u001b[0;36mBaseAsyncIOLoop.start\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 213\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m 214\u001b[0m asyncio\u001b[38;5;241m.\u001b[39mset_event_loop(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39masyncio_loop)\n\u001b[0;32m--> 215\u001b[0m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43masyncio_loop\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun_forever\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 216\u001b[0m \u001b[38;5;28;01mfinally\u001b[39;00m:\n\u001b[1;32m 217\u001b[0m asyncio\u001b[38;5;241m.\u001b[39mset_event_loop(old_loop)\n",
+ "File \u001b[0;32m/opt/conda/lib/python3.9/site-packages/nest_asyncio.py:73\u001b[0m, in \u001b[0;36m_patch_loop.<locals>.run_forever\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 71\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m manage_run(\u001b[38;5;28mself\u001b[39m), manage_asyncgens(\u001b[38;5;28mself\u001b[39m):\n\u001b[1;32m 72\u001b[0m \u001b[38;5;28;01mwhile\u001b[39;00m \u001b[38;5;28;01mTrue\u001b[39;00m:\n\u001b[0;32m---> 73\u001b[0m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_run_once\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 74\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_stopping:\n\u001b[1;32m 75\u001b[0m \u001b[38;5;28;01mbreak\u001b[39;00m\n",
+ "File \u001b[0;32m/opt/conda/lib/python3.9/site-packages/nest_asyncio.py:107\u001b[0m, in \u001b[0;36m_patch_loop.<locals>._run_once\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 100\u001b[0m heappop(scheduled)\n\u001b[1;32m 102\u001b[0m timeout \u001b[38;5;241m=\u001b[39m (\n\u001b[1;32m 103\u001b[0m \u001b[38;5;241m0\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m ready \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_stopping\n\u001b[1;32m 104\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28mmin\u001b[39m(\u001b[38;5;28mmax\u001b[39m(\n\u001b[1;32m 105\u001b[0m scheduled[\u001b[38;5;241m0\u001b[39m]\u001b[38;5;241m.\u001b[39m_when \u001b[38;5;241m-\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtime(), \u001b[38;5;241m0\u001b[39m), \u001b[38;5;241m86400\u001b[39m) \u001b[38;5;28;01mif\u001b[39;00m scheduled\n\u001b[1;32m 106\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m)\n\u001b[0;32m--> 107\u001b[0m event_list \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_selector\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mselect\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtimeout\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 108\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_process_events(event_list)\n\u001b[1;32m 110\u001b[0m end_time \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtime() \u001b[38;5;241m+\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_clock_resolution\n",
+ "File \u001b[0;32m/opt/conda/lib/python3.9/selectors.py:469\u001b[0m, in \u001b[0;36mEpollSelector.select\u001b[0;34m(self, timeout)\u001b[0m\n\u001b[1;32m 467\u001b[0m ready \u001b[38;5;241m=\u001b[39m []\n\u001b[1;32m 468\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 469\u001b[0m fd_event_list \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_selector\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpoll\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtimeout\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmax_ev\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 470\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mInterruptedError\u001b[39;00m:\n\u001b[1;32m 471\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m ready\n",
+ "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
+ ]
+ }
+ ],
+ "source": [
+ "from nuplan.planning.script.run_nuboard import main as main_nuboard\n",
+ "\n",
+ "# Run nuBoard\n",
+ "main_nuboard(cfg)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 21,
+ "id": "d966448b",
+ "metadata": {
+ "scrolled": true
+ },
+ "outputs": [
+ {
+ "ename": "NameError",
+ "evalue": "name 'runner' is not defined",
+ "output_type": "error",
+ "traceback": [
+ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+ "\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
+ "Cell \u001b[0;32mIn[21], line 2\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtutorials\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutils\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mtutorial_utils\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m visualize_history\n\u001b[0;32m----> 2\u001b[0m visualize_history(\u001b[43mrunner\u001b[49m\u001b[38;5;241m.\u001b[39msimulation\u001b[38;5;241m.\u001b[39m_history, runner\u001b[38;5;241m.\u001b[39mscenario, bokeh_port\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m5007\u001b[39m)\n",
+ "\u001b[0;31mNameError\u001b[0m: name 'runner' is not defined"
+ ]
+ }
+ ],
+ "source": [
+ "from tutorials.utils.tutorial_utils import visualize_history\n",
+ "visualize_history(runner.simulation._history, runner.scenario, bokeh_port=5007)"
+ ]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "1bdb7bef",
- "metadata": {},
+ "id": "4d3f0ce6",
+ "metadata": {
+ "scrolled": true
+ },
"outputs": [],
"source": []
}
diff --git a/experiments/test_notebook.ipynb b/experiments/test_notebook.ipynb
index bac465598e8943cf46f8651c520ecc9a6c7ebd46..d98819ecf99b52ab0ec1e1a40c9ca869999a1cc5 100644
--- a/experiments/test_notebook.ipynb
+++ b/experiments/test_notebook.ipynb
@@ -26,7 +26,7 @@
"name": "stderr",
"output_type": "stream",
"text": [
- "/tmp/ipykernel_2455/4095267831.py:5: DeprecationWarning: Importing display from IPython.core.display is deprecated since IPython 7.14, please import from IPython display\n",
+ "/tmp/ipykernel_3875/4095267831.py:5: DeprecationWarning: Importing display from IPython.core.display is deprecated since IPython 7.14, please import from IPython display\n",
" from IPython.core.display import display, HTML\n"
]
}
@@ -436,10 +436,18 @@
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": 12,
"id": "11b08c6d",
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "default_simulation\n"
+ ]
+ }
+ ],
"source": [
"from tutorials.utils.tutorial_utils import construct_simulation_hydra_paths\n",
"\n",
@@ -464,6 +472,7 @@
"hydra.initialize(config_path=simulation_hydra_paths.config_path)\n",
"\n",
"# Compose the configuration\n",
+ "print(simulation_hydra_paths.config_name)\n",
"cfg = hydra.compose(config_name=simulation_hydra_paths.config_name, overrides=[\n",
" '+simulation=closed_loop_reactive_agents',\n",
" #'model=pgm_hybrid_model',\n",
@@ -479,7 +488,9 @@
" '+occlusion.manager_type=wedge', #options: [range, shadow, wedge]\n",
" \"hydra.searchpath=[pkg://tuplan_garage.planning.script.config.common, pkg://tuplan_garage.planning.script.config.simulation, pkg://nuplan.planning.script.config.common, pkg://nuplan.planning.script.experiments]\",\n",
" *DATASET_PARAMS,\n",
- "])"
+ "])\n",
+ "\n",
+ "output_folder = cfg.output_dir"
]
},
{
@@ -511,32 +522,32 @@
"name": "stdout",
"output_type": "stream",
"text": [
- "2023-12-04 23:36:26,261 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:19} Building WorkerPool...\n",
- "2023-12-04 23:36:26,263 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:101} Worker: Sequential\n",
- "2023-12-04 23:36:26,263 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:102} Number of nodes: 1\n",
+ "2023-12-06 01:45:40,942 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:19} Building WorkerPool...\n",
+ "2023-12-06 01:45:40,943 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:101} Worker: Sequential\n",
+ "2023-12-06 01:45:40,943 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/utils/multithreading/worker_pool.py:102} Number of nodes: 1\n",
"Number of CPUs per node: 1\n",
"Number of GPUs per node: 0\n",
"Number of threads across all nodes: 1\n",
- "2023-12-04 23:36:26,263 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:27} Building WorkerPool...DONE!\n",
- "2023-12-04 23:36:26,263 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:32} Building experiment folders...\n",
- "2023-12-04 23:36:26,263 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:35} \n",
+ "2023-12-06 01:45:40,943 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/worker_pool_builder.py:27} Building WorkerPool...DONE!\n",
+ "2023-12-06 01:45:40,943 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:32} Building experiment folders...\n",
+ "2023-12-06 01:45:40,943 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:35} \n",
"\n",
- "\tFolder where all results are stored: /home/ehdykhne/Repos/Datasets/nuplan/exp/exp/simulation/closed_loop_reactive_agents/2023.12.04.23.36.25\n",
+ "\tFolder where all results are stored: /home/ehdykhne/Repos/Datasets/nuplan/exp/exp/simulation/closed_loop_reactive_agents/2023.12.06.01.45.39\n",
"\n",
- "2023-12-04 23:36:26,293 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:70} Building experiment folders...DONE!\n",
- "2023-12-04 23:36:26,293 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:52} Building AbstractCallback...\n",
- "2023-12-04 23:36:26,293 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:68} Building AbstractCallback: 0...DONE!\n",
- "2023-12-04 23:36:26,293 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:49} Building simulations...\n",
- "2023-12-04 23:36:26,293 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:55} Extracting scenarios...\n",
- "2023-12-04 23:36:26,294 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/common/utils/distributed_scenario_filter.py:83} Building Scenarios in mode DistributedMode.SINGLE_NODE\n",
- "2023-12-04 23:36:26,294 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:18} Building AbstractScenarioBuilder...\n",
- "2023-12-04 23:36:26,305 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:21} Building AbstractScenarioBuilder...DONE!\n",
- "2023-12-04 23:36:26,305 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:35} Building ScenarioFilter...\n",
- "2023-12-04 23:36:26,306 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:44} Building ScenarioFilter...DONE!\n",
- "2023-12-04 23:36:26,381 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:76} Building metric engines...\n",
- "2023-12-04 23:36:26,403 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:78} Building metric engines...DONE\n",
- "2023-12-04 23:36:26,403 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:82} Building simulations from 1 scenarios...\n",
- "2023-12-04 23:36:26,979 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:142} Building simulations...DONE!\n"
+ "2023-12-06 01:45:40,945 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/folder_builder.py:70} Building experiment folders...DONE!\n",
+ "2023-12-06 01:45:40,945 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:52} Building AbstractCallback...\n",
+ "2023-12-06 01:45:40,945 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_callback_builder.py:68} Building AbstractCallback: 0...DONE!\n",
+ "2023-12-06 01:45:40,945 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:49} Building simulations...\n",
+ "2023-12-06 01:45:40,945 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:55} Extracting scenarios...\n",
+ "2023-12-06 01:45:40,945 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/common/utils/distributed_scenario_filter.py:83} Building Scenarios in mode DistributedMode.SINGLE_NODE\n",
+ "2023-12-06 01:45:40,946 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:18} Building AbstractScenarioBuilder...\n",
+ "2023-12-06 01:45:40,956 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:21} Building AbstractScenarioBuilder...DONE!\n",
+ "2023-12-06 01:45:40,956 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:35} Building ScenarioFilter...\n",
+ "2023-12-06 01:45:40,957 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_filter_builder.py:44} Building ScenarioFilter...DONE!\n",
+ "2023-12-06 01:45:41,031 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:76} Building metric engines...\n",
+ "2023-12-06 01:45:41,058 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:78} Building metric engines...DONE\n",
+ "2023-12-06 01:45:41,058 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:82} Building simulations from 1 scenarios...\n",
+ "2023-12-06 01:45:41,608 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/simulation_builder.py:142} Building simulations...DONE!\n"
]
}
],
@@ -562,7 +573,9 @@
"cell_type": "code",
"execution_count": 8,
"id": "90b79421",
- "metadata": {},
+ "metadata": {
+ "scrolled": true
+ },
"outputs": [
{
"name": "stderr",
@@ -570,182 +583,6 @@
"text": [
" \r"
]
- },
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "elapsed time: 0.028219938278198242\n",
- "elapsed time: 0.02777719497680664\n",
- "elapsed time: 0.02846693992614746\n",
- "elapsed time: 0.0270230770111084\n",
- "elapsed time: 0.027086973190307617\n",
- "elapsed time: 0.02704000473022461\n",
- "elapsed time: 0.02687692642211914\n",
- "elapsed time: 0.02667236328125\n",
- "elapsed time: 0.027269363403320312\n",
- "elapsed time: 0.026731491088867188\n",
- "elapsed time: 0.02642965316772461\n",
- "elapsed time: 0.02652263641357422\n",
- "elapsed time: 0.026285171508789062\n",
- "elapsed time: 0.026721715927124023\n",
- "elapsed time: 0.02680516242980957\n",
- "elapsed time: 0.026640892028808594\n",
- "elapsed time: 0.02759575843811035\n",
- "elapsed time: 0.02881908416748047\n",
- "elapsed time: 0.02743077278137207\n",
- "elapsed time: 0.02760910987854004\n",
- "elapsed time: 0.026903390884399414\n",
- "elapsed time: 0.027498245239257812\n",
- "elapsed time: 0.02749776840209961\n",
- "elapsed time: 0.027604103088378906\n",
- "elapsed time: 0.028120994567871094\n",
- "elapsed time: 0.028057098388671875\n",
- "elapsed time: 0.02862381935119629\n",
- "elapsed time: 0.02895808219909668\n",
- "elapsed time: 0.02965092658996582\n",
- "elapsed time: 0.02978205680847168\n",
- "elapsed time: 0.030011653900146484\n",
- "elapsed time: 0.029865503311157227\n",
- "elapsed time: 0.03024911880493164\n",
- "elapsed time: 0.02980661392211914\n",
- "elapsed time: 0.030503511428833008\n",
- "elapsed time: 0.0301516056060791\n",
- "elapsed time: 0.0298004150390625\n",
- "elapsed time: 0.029525279998779297\n",
- "elapsed time: 0.029907703399658203\n",
- "elapsed time: 0.029463529586791992\n",
- "elapsed time: 0.02914571762084961\n",
- "elapsed time: 0.0291597843170166\n",
- "elapsed time: 0.02936530113220215\n",
- "elapsed time: 0.029200077056884766\n",
- "elapsed time: 0.0286557674407959\n",
- "elapsed time: 0.02902507781982422\n",
- "elapsed time: 0.028758764266967773\n",
- "elapsed time: 0.02935171127319336\n",
- "elapsed time: 0.029259204864501953\n",
- "elapsed time: 0.029033422470092773\n",
- "elapsed time: 0.028880834579467773\n",
- "elapsed time: 0.028924226760864258\n",
- "elapsed time: 0.028826475143432617\n",
- "elapsed time: 0.028478622436523438\n",
- "elapsed time: 0.0288240909576416\n",
- "elapsed time: 0.02877044677734375\n",
- "elapsed time: 0.028140544891357422\n",
- "elapsed time: 0.02760028839111328\n",
- "elapsed time: 0.027898073196411133\n",
- "elapsed time: 0.02771472930908203\n",
- "elapsed time: 0.027673721313476562\n",
- "elapsed time: 0.02758169174194336\n",
- "elapsed time: 0.027628183364868164\n",
- "elapsed time: 0.027477741241455078\n",
- "elapsed time: 0.027070999145507812\n",
- "elapsed time: 0.026733875274658203\n",
- "elapsed time: 0.02656245231628418\n",
- "elapsed time: 0.026247739791870117\n",
- "elapsed time: 0.02589249610900879\n",
- "elapsed time: 0.025628089904785156\n",
- "elapsed time: 0.025936126708984375\n",
- "elapsed time: 0.0257565975189209\n",
- "elapsed time: 0.025776147842407227\n",
- "elapsed time: 0.025105953216552734\n",
- "elapsed time: 0.024316072463989258\n",
- "elapsed time: 0.02410602569580078\n",
- "elapsed time: 0.0240628719329834\n",
- "elapsed time: 0.023929595947265625\n",
- "elapsed time: 0.024273395538330078\n",
- "elapsed time: 0.023852109909057617\n",
- "elapsed time: 0.023021221160888672\n",
- "elapsed time: 0.024117708206176758\n",
- "elapsed time: 0.023128271102905273\n",
- "elapsed time: 0.02261972427368164\n",
- "elapsed time: 0.022465229034423828\n",
- "elapsed time: 0.022220611572265625\n",
- "elapsed time: 0.021961688995361328\n",
- "elapsed time: 0.02196502685546875\n",
- "elapsed time: 0.021490097045898438\n",
- "elapsed time: 0.02141594886779785\n",
- "elapsed time: 0.02118968963623047\n",
- "elapsed time: 0.021246671676635742\n",
- "elapsed time: 0.020877838134765625\n",
- "elapsed time: 0.020760774612426758\n",
- "elapsed time: 0.020011425018310547\n",
- "elapsed time: 0.020204544067382812\n",
- "elapsed time: 0.019935131072998047\n",
- "elapsed time: 0.020002126693725586\n",
- "elapsed time: 0.020210981369018555\n",
- "elapsed time: 0.01934981346130371\n",
- "elapsed time: 0.0188140869140625\n",
- "elapsed time: 0.01867389678955078\n",
- "elapsed time: 0.018260478973388672\n",
- "elapsed time: 0.01837778091430664\n",
- "elapsed time: 0.018278837203979492\n",
- "elapsed time: 0.018478870391845703\n",
- "elapsed time: 0.018354177474975586\n",
- "elapsed time: 0.01824164390563965\n",
- "elapsed time: 0.017989397048950195\n",
- "elapsed time: 0.018170833587646484\n",
- "elapsed time: 0.017641782760620117\n",
- "elapsed time: 0.017865657806396484\n",
- "elapsed time: 0.017589569091796875\n",
- "elapsed time: 0.01693439483642578\n",
- "elapsed time: 0.016333580017089844\n",
- "elapsed time: 0.016154050827026367\n",
- "elapsed time: 0.01594376564025879\n",
- "elapsed time: 0.015879392623901367\n",
- "elapsed time: 0.015716075897216797\n",
- "elapsed time: 0.01625823974609375\n",
- "elapsed time: 0.015722990036010742\n",
- "elapsed time: 0.01554727554321289\n",
- "elapsed time: 0.015479564666748047\n",
- "elapsed time: 0.01556706428527832\n",
- "elapsed time: 0.015661954879760742\n",
- "elapsed time: 0.01565694808959961\n",
- "elapsed time: 0.014882326126098633\n",
- "elapsed time: 0.014642000198364258\n",
- "elapsed time: 0.014568328857421875\n",
- "elapsed time: 0.014591693878173828\n",
- "elapsed time: 0.014154195785522461\n",
- "elapsed time: 0.014048576354980469\n",
- "elapsed time: 0.013832807540893555\n",
- "elapsed time: 0.013761281967163086\n",
- "elapsed time: 0.013667583465576172\n",
- "elapsed time: 0.01350092887878418\n",
- "elapsed time: 0.013176918029785156\n",
- "elapsed time: 0.013062000274658203\n",
- "elapsed time: 0.013159036636352539\n",
- "elapsed time: 0.013329505920410156\n",
- "elapsed time: 0.013332366943359375\n",
- "elapsed time: 0.013476133346557617\n",
- "elapsed time: 0.013322591781616211\n",
- "elapsed time: 0.01374506950378418\n",
- "elapsed time: 0.013417720794677734\n",
- "elapsed time: 0.013437509536743164\n",
- "elapsed time: 0.013459444046020508\n",
- "elapsed time: 0.013498783111572266\n",
- "elapsed time: 0.013631343841552734\n",
- "elapsed time: 0.013602733612060547\n",
- "elapsed time: 0.013532161712646484\n",
- "elapsed time: 0.013321399688720703\n",
- "elapsed time: 0.013411521911621094\n",
- "elapsed time: 0.013401508331298828\n",
- "elapsed time: 0.013605833053588867\n",
- "elapsed time: 0.013462543487548828\n",
- "elapsed time: 0.013318061828613281\n",
- "elapsed time: 0.013405323028564453\n",
- "elapsed time: 0.013209819793701172\n",
- "elapsed time: 0.013265848159790039\n",
- "elapsed time: 0.01320028305053711\n",
- "elapsed time: 0.013041496276855469\n",
- "elapsed time: 0.013081550598144531\n",
- "elapsed time: 0.013251304626464844\n",
- "elapsed time: 0.0133819580078125\n",
- "elapsed time: 0.013030290603637695\n",
- "elapsed time: 0.013294696807861328\n",
- "elapsed time: 0.013309478759765625\n",
- "elapsed time: 0.013379335403442383\n",
- "elapsed time: 0.01327824592590332\n"
- ]
}
],
"source": [
@@ -805,7 +642,7 @@
" (function() {\n",
" const xhr = new XMLHttpRequest()\n",
" xhr.responseType = 'blob';\n",
- " xhr.open('GET', \"http://localhost:5006/autoload.js?bokeh-autoload-element=1003&bokeh-absolute-url=http://localhost:5006&resources=none\", true);\n",
+ " xhr.open('GET', \"http://localhost:5007/autoload.js?bokeh-autoload-element=1003&bokeh-absolute-url=http://localhost:5007&resources=none\", true);\n",
" xhr.onload = function (event) {\n",
" const script = document.createElement('script');\n",
" const src = URL.createObjectURL(event.target.response);\n",
@@ -819,29 +656,92 @@
},
"metadata": {
"application/vnd.bokehjs_exec.v0+json": {
- "server_id": "e6e966ebbd494438b1ac64effae4c8b2"
+ "server_id": "96025b6f43524045a939a4221ccc3af1"
}
},
"output_type": "display_data"
+ }
+ ],
+ "source": [
+ "from tutorials.utils.tutorial_utils import visualize_history\n",
+ "visualize_history(runner.simulation._history, runner.scenario, bokeh_port=5007)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "ee650f8f",
+ "metadata": {},
+ "source": [
+ "## Prepare the nuBoard config"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "id": "887a51e2",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "# Location of path with all nuBoard configs\n",
+ "CONFIG_PATH = '../nuplan/planning/script/config/nuboard'\n",
+ "CONFIG_NAME = 'default_nuboard'\n",
+ "\n",
+ "# Initialize configuration management system\n",
+ "hydra.core.global_hydra.GlobalHydra.instance().clear() # reinitialize hydra if already initialized\n",
+ "hydra.initialize(config_path=CONFIG_PATH)\n",
+ "\n",
+ "# Compose the configuration\n",
+ "cfg = hydra.compose(config_name=CONFIG_NAME, overrides=[\n",
+ " 'scenario_builder=nuplan_mini', # set the database (same as simulation) used to fetch data for visualization\n",
+ " f'simulation_path={[output_folder]}', # nuboard file path(s), if left empty the user can open the file inside nuBoard\n",
+ "])\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "id": "f5149ddd",
+ "metadata": {},
+ "source": [
+ "## Launch nuBoard (open in new tab - recommended)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "id": "67b67b86",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "INFO:bokeh.server.server:Starting Bokeh server version 2.4.3 (running on Tornado 6.2)\n",
+ "WARNING:bokeh.server.util:Host wildcard '*' will allow connections originating from multiple (or possibly all) hostnames or IPs. Use non-wildcard values to restrict access explicitly\n",
+ "INFO:bokeh.server.tornado:User authentication hooks NOT provided (default user enabled)\n"
+ ]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "2023-12-04 23:37:42,023 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
- "2023-12-04 23:37:42,024 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
- "2023-12-04 23:37:42,026 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Error creating dataset. Could not read schema from 'pretrained_checkpoints/gc_pgp_checkpoint.ckpt'. Is this a 'parquet' file?: Could not open Parquet input source 'pretrained_checkpoints/gc_pgp_checkpoint.ckpt': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
- "2023-12-04 23:37:42,027 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Error creating dataset. Could not read schema from '.ipynb_checkpoints/test_notebook-checkpoint.ipynb'. Is this a 'parquet' file?: Could not open Parquet input source '.ipynb_checkpoints/test_notebook-checkpoint.ipynb': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
- "2023-12-04 23:37:42,028 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Error creating dataset. Could not read schema from 'run_sim_closed_loop/training_raster_experiment/train_default_raster/2023.11.14.22.55.23/hparams.yaml'. Is this a 'parquet' file?: Could not open Parquet input source 'run_sim_closed_loop/training_raster_experiment/train_default_raster/2023.11.14.22.55.23/hparams.yaml': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
- "2023-12-04 23:37:42,028 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
- "2023-12-04 23:37:42,032 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/simulation_tile.py:172} Minimum frame time=0.017 s\n"
+ "2023-12-06 01:46:42,603 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:18} Building AbstractScenarioBuilder...\n",
+ "2023-12-06 01:46:42,614 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/script/builders/scenario_building_builder.py:21} Building AbstractScenarioBuilder...DONE!\n",
+ "2023-12-06 01:46:42,615 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/nuboard.py:84} Opening Bokeh application on http://localhost:5006/\n",
+ "2023-12-06 01:46:42,615 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/nuboard.py:85} Async rendering is set to: True\n",
+ "2023-12-06 01:47:06,044 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:06,045 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:06,045 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Error creating dataset. Could not read schema from 'pretrained_checkpoints/gc_pgp_checkpoint.ckpt'. Is this a 'parquet' file?: Could not open Parquet input source 'pretrained_checkpoints/gc_pgp_checkpoint.ckpt': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:06,045 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Error creating dataset. Could not read schema from '.ipynb_checkpoints/test_notebook-checkpoint.ipynb'. Is this a 'parquet' file?: Could not open Parquet input source '.ipynb_checkpoints/test_notebook-checkpoint.ipynb': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:06,046 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Error creating dataset. Could not read schema from 'run_sim_closed_loop/training_raster_experiment/train_default_raster/2023.11.14.22.55.23/hparams.yaml'. Is this a 'parquet' file?: Could not open Parquet input source 'run_sim_closed_loop/training_raster_experiment/train_default_raster/2023.11.14.22.55.23/hparams.yaml': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:06,046 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:06,047 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/simulation_tile.py:172} Minimum frame time=0.017 s\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
- "Rendering a scenario: 100%|██████████| 1/1 [00:00<00:00, 157.64it/s]\n",
+ "Rendering a scenario: 100%|██████████| 1/1 [00:00<00:00, 152.68it/s]\n",
"WARNING:bokeh.core.validation.check:W-1000 (MISSING_RENDERERS): Plot has no renderers: Figure(id='1005', ...)\n"
]
},
@@ -849,99 +749,107 @@
"name": "stdout",
"output_type": "stream",
"text": [
- "2023-12-04 23:37:44,012 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET / (10.40.117.44) 2030.70ms\n",
- "2023-12-04 23:37:44,023 INFO {/home/ehdykhne/Repos/nuplan-devkit/tutorials/utils/tutorial_utils.py:267} Done rendering!\n"
+ "2023-12-06 01:47:06,996 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET / (10.40.117.44) 962.39ms\n",
+ "2023-12-06 01:47:06,997 INFO {/home/ehdykhne/Repos/nuplan-devkit/tutorials/utils/tutorial_utils.py:267} Done rendering!\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
- "INFO:tornado.access:200 GET / (10.40.117.44) 2030.70ms\n"
+ "INFO:tornado.access:200 GET / (10.40.117.44) 962.39ms\n",
+ "INFO:bokeh.server.views.ws:WebSocket connection opened\n",
+ "INFO:tornado.access:101 GET /ws (10.40.117.44) 0.48ms\n",
+ "INFO:bokeh.server.views.ws:ServerConnection created\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "2023-12-04 23:37:44,370 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET /static/js/bokeh-gl.min.js?v=e5df31fd9010eacff0aa72d315264604b5e34972ba445acea6fce98080eecf33acf2d2986126360faaa5852813cffa16f6f6f4889923318300f062497c02da4e (10.40.117.44) 271.87ms\n",
- "2023-12-04 23:37:44,398 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET /static/js/bokeh.min.js?v=3c61e952b808bb7e346ce828a565a5f23aaf7708d034fa9d0906403813355d45bb4e8d8b0b23a93f032c76831d4f0221846f28699c7f5147caa62e0d31668314 (10.40.117.44) 299.84ms\n",
- "2023-12-04 23:37:44,400 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET /static/js/bokeh-widgets.min.js?v=8a1ff6f5aa0d967f4998d275803bbb111d928fd9f605ef9e1f30cfd021df0e77224ee3d13f83edb3a942f6e4ccc569ee5dd8951a8aa6cb600602463b90c65a87 (10.40.117.44) 301.22ms\n",
- "2023-12-04 23:37:44,435 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET /static/js/bokeh-tables.min.js?v=ae2903e57cf57f52819fdf4d938c648982b51c34f73b6e653a0f3bb3c8ab44f338505931ace43eafc1636e215492e2314acf54c54baffb47813b86b4923a7fe0 (10.40.117.44) 37.66ms\n"
+ "2023-12-06 01:47:07,419 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 101 GET /ws (10.40.117.44) 0.48ms\n",
+ "2023-12-06 01:47:07,942 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:08,002 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:08,013 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Error creating dataset. Could not read schema from 'pretrained_checkpoints/gc_pgp_checkpoint.ckpt'. Is this a 'parquet' file?: Could not open Parquet input source 'pretrained_checkpoints/gc_pgp_checkpoint.ckpt': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:08,038 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Error creating dataset. Could not read schema from '.ipynb_checkpoints/test_notebook-checkpoint.ipynb'. Is this a 'parquet' file?: Could not open Parquet input source '.ipynb_checkpoints/test_notebook-checkpoint.ipynb': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:08,054 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Error creating dataset. Could not read schema from 'run_sim_closed_loop/training_raster_experiment/train_default_raster/2023.11.14.22.55.23/hparams.yaml'. Is this a 'parquet' file?: Could not open Parquet input source 'run_sim_closed_loop/training_raster_experiment/train_default_raster/2023.11.14.22.55.23/hparams.yaml': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:08,096 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:08,476 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/simulation_tile.py:172} Minimum frame time=0.017 s\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
- "INFO:tornado.access:200 GET /static/js/bokeh-gl.min.js?v=e5df31fd9010eacff0aa72d315264604b5e34972ba445acea6fce98080eecf33acf2d2986126360faaa5852813cffa16f6f6f4889923318300f062497c02da4e (10.40.117.44) 271.87ms\n",
- "INFO:tornado.access:200 GET /static/js/bokeh.min.js?v=3c61e952b808bb7e346ce828a565a5f23aaf7708d034fa9d0906403813355d45bb4e8d8b0b23a93f032c76831d4f0221846f28699c7f5147caa62e0d31668314 (10.40.117.44) 299.84ms\n",
- "INFO:tornado.access:200 GET /static/js/bokeh-widgets.min.js?v=8a1ff6f5aa0d967f4998d275803bbb111d928fd9f605ef9e1f30cfd021df0e77224ee3d13f83edb3a942f6e4ccc569ee5dd8951a8aa6cb600602463b90c65a87 (10.40.117.44) 301.22ms\n",
- "INFO:tornado.access:200 GET /static/js/bokeh-tables.min.js?v=ae2903e57cf57f52819fdf4d938c648982b51c34f73b6e653a0f3bb3c8ab44f338505931ace43eafc1636e215492e2314acf54c54baffb47813b86b4923a7fe0 (10.40.117.44) 37.66ms\n",
- "INFO:tornado.access:200 GET /static/js/bokeh-mathjax.min.js?v=176c36fdbcd8fc1019fc828101a2804081a35baf4018d7f2633cd263156b593aa73112f400112b662daa0590138b74851bc91f1f2a5fbf5416ee8c876c3e0d0c (10.40.117.44) 135.59ms\n",
- "INFO:bokeh.server.views.ws:WebSocket connection opened\n",
- "INFO:tornado.access:101 GET /ws (10.40.117.44) 1.63ms\n",
- "INFO:bokeh.server.views.ws:ServerConnection created\n"
+ "Rendering a scenario: 100%|██████████| 1/1 [00:00<00:00, 46.27it/s]\n",
+ "WARNING:bokeh.core.validation.check:W-1000 (MISSING_RENDERERS): Plot has no renderers: Figure(id='3457', ...)\n",
+ "INFO:tornado.access:200 GET / (10.40.117.44) 1694.56ms\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "2023-12-04 23:37:45,376 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET /static/js/bokeh-mathjax.min.js?v=176c36fdbcd8fc1019fc828101a2804081a35baf4018d7f2633cd263156b593aa73112f400112b662daa0590138b74851bc91f1f2a5fbf5416ee8c876c3e0d0c (10.40.117.44) 135.59ms\n",
- "2023-12-04 23:37:45,551 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 101 GET /ws (10.40.117.44) 1.63ms\n",
- "2023-12-04 23:37:45,664 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
- "2023-12-04 23:37:45,666 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
- "2023-12-04 23:37:45,666 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Error creating dataset. Could not read schema from 'pretrained_checkpoints/gc_pgp_checkpoint.ckpt'. Is this a 'parquet' file?: Could not open Parquet input source 'pretrained_checkpoints/gc_pgp_checkpoint.ckpt': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
- "2023-12-04 23:37:45,667 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Error creating dataset. Could not read schema from '.ipynb_checkpoints/test_notebook-checkpoint.ipynb'. Is this a 'parquet' file?: Could not open Parquet input source '.ipynb_checkpoints/test_notebook-checkpoint.ipynb': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
- "2023-12-04 23:37:45,667 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Error creating dataset. Could not read schema from 'run_sim_closed_loop/training_raster_experiment/train_default_raster/2023.11.14.22.55.23/hparams.yaml'. Is this a 'parquet' file?: Could not open Parquet input source 'run_sim_closed_loop/training_raster_experiment/train_default_raster/2023.11.14.22.55.23/hparams.yaml': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
- "2023-12-04 23:37:45,667 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
- "2023-12-04 23:37:45,669 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/simulation_tile.py:172} Minimum frame time=0.017 s\n"
+ "2023-12-06 01:47:09,573 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET / (10.40.117.44) 1694.56ms\n",
+ "2023-12-06 01:47:11,850 INFO {/home/ehdykhne/Repos/nuplan-devkit/tutorials/utils/tutorial_utils.py:267} Done rendering!\n",
+ "2023-12-06 01:47:11,858 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:11,864 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/simulation_tile.py:172} Minimum frame time=0.017 s\n",
+ "2023-12-06 01:47:11,867 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/tabs/scenario_tab.py:485} Rending scenario plot takes 0.0005 seconds.\n",
+ "2023-12-06 01:47:11,970 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET / (10.40.117.44) 118.54ms\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
- "Rendering a scenario: 100%|██████████| 1/1 [00:00<00:00, 10.45it/s]\n",
- "WARNING:bokeh.core.validation.check:W-1000 (MISSING_RENDERERS): Plot has no renderers: Figure(id='3857', ...)\n",
- "INFO:tornado.access:200 GET / (10.40.117.44) 805.62ms\n"
+ "INFO:tornado.access:200 GET / (10.40.117.44) 118.54ms\n",
+ "INFO:tornado.access:200 GET / (10.40.117.44) 114.48ms\n",
+ "INFO:bokeh.server.views.ws:WebSocket connection opened\n",
+ "INFO:tornado.access:101 GET /ws (10.40.117.44) 9.93ms\n",
+ "INFO:bokeh.server.views.ws:ServerConnection created\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
- "2023-12-04 23:37:46,467 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET / (10.40.117.44) 805.62ms\n",
- "2023-12-04 23:37:46,513 INFO {/home/ehdykhne/Repos/nuplan-devkit/tutorials/utils/tutorial_utils.py:267} Done rendering!\n"
+ "2023-12-06 01:47:12,984 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/experiment_file_data.py:140} Could not open Parquet input source '<Buffer>': Parquet magic bytes not found in footer. Either the file is corrupted or this is not a parquet file.\n",
+ "2023-12-06 01:47:12,989 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/base/simulation_tile.py:172} Minimum frame time=0.017 s\n",
+ "2023-12-06 01:47:12,992 INFO {/home/ehdykhne/Repos/nuplan-devkit/nuplan/planning/nuboard/tabs/scenario_tab.py:485} Rending scenario plot takes 0.0005 seconds.\n",
+ "2023-12-06 01:47:13,093 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 200 GET / (10.40.117.44) 114.48ms\n",
+ "2023-12-06 01:47:13,103 INFO {/opt/conda/lib/python3.9/site-packages/tornado/web.py:2271} 101 GET /ws (10.40.117.44) 9.93ms\n"
+ ]
+ },
+ {
+ "ename": "KeyboardInterrupt",
+ "evalue": "",
+ "output_type": "error",
+ "traceback": [
+ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+ "\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)",
+ "Cell \u001b[0;32mIn[11], line 4\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mnuplan\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mplanning\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mscript\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mrun_nuboard\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m main \u001b[38;5;28;01mas\u001b[39;00m main_nuboard\n\u001b[1;32m 3\u001b[0m \u001b[38;5;66;03m# Run nuBoard\u001b[39;00m\n\u001b[0;32m----> 4\u001b[0m \u001b[43mmain_nuboard\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcfg\u001b[49m\u001b[43m)\u001b[49m\n",
+ "File \u001b[0;32m/opt/conda/lib/python3.9/site-packages/hydra/main.py:44\u001b[0m, in \u001b[0;36mmain.<locals>.main_decorator.<locals>.decorated_main\u001b[0;34m(cfg_passthrough)\u001b[0m\n\u001b[1;32m 41\u001b[0m \u001b[38;5;129m@functools\u001b[39m\u001b[38;5;241m.\u001b[39mwraps(task_function)\n\u001b[1;32m 42\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mdecorated_main\u001b[39m(cfg_passthrough: Optional[DictConfig] \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m Any:\n\u001b[1;32m 43\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m cfg_passthrough \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m---> 44\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mtask_function\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcfg_passthrough\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 45\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m 46\u001b[0m args \u001b[38;5;241m=\u001b[39m get_args_parser()\n",
+ "File \u001b[0;32m~/Repos/nuplan-devkit/nuplan/planning/script/run_nuboard.py:74\u001b[0m, in \u001b[0;36mmain\u001b[0;34m(cfg)\u001b[0m\n\u001b[1;32m 69\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m 70\u001b[0m \u001b[38;5;124;03mExecute all available challenges simultaneously on the same scenario.\u001b[39;00m\n\u001b[1;32m 71\u001b[0m \u001b[38;5;124;03m:param cfg: DictConfig. Configuration that is used to run the experiment.\u001b[39;00m\n\u001b[1;32m 72\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m 73\u001b[0m nuboard \u001b[38;5;241m=\u001b[39m initialize_nuboard(cfg)\n\u001b[0;32m---> 74\u001b[0m \u001b[43mnuboard\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n",
+ "File \u001b[0;32m~/Repos/nuplan-devkit/nuplan/planning/nuboard/nuboard.py:112\u001b[0m, in \u001b[0;36mNuBoard.run\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 110\u001b[0m \u001b[38;5;66;03m# Catch RuntimeError in jupyter notebook\u001b[39;00m\n\u001b[1;32m 111\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 112\u001b[0m \u001b[43mio_loop\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mstart\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 113\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mRuntimeError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m 114\u001b[0m logger\u001b[38;5;241m.\u001b[39mwarning(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00me\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n",
+ "File \u001b[0;32m/opt/conda/lib/python3.9/site-packages/tornado/platform/asyncio.py:215\u001b[0m, in \u001b[0;36mBaseAsyncIOLoop.start\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 213\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m 214\u001b[0m asyncio\u001b[38;5;241m.\u001b[39mset_event_loop(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39masyncio_loop)\n\u001b[0;32m--> 215\u001b[0m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43masyncio_loop\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun_forever\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 216\u001b[0m \u001b[38;5;28;01mfinally\u001b[39;00m:\n\u001b[1;32m 217\u001b[0m asyncio\u001b[38;5;241m.\u001b[39mset_event_loop(old_loop)\n",
+ "File \u001b[0;32m/opt/conda/lib/python3.9/site-packages/nest_asyncio.py:73\u001b[0m, in \u001b[0;36m_patch_loop.<locals>.run_forever\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 71\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m manage_run(\u001b[38;5;28mself\u001b[39m), manage_asyncgens(\u001b[38;5;28mself\u001b[39m):\n\u001b[1;32m 72\u001b[0m \u001b[38;5;28;01mwhile\u001b[39;00m \u001b[38;5;28;01mTrue\u001b[39;00m:\n\u001b[0;32m---> 73\u001b[0m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_run_once\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 74\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_stopping:\n\u001b[1;32m 75\u001b[0m \u001b[38;5;28;01mbreak\u001b[39;00m\n",
+ "File \u001b[0;32m/opt/conda/lib/python3.9/site-packages/nest_asyncio.py:107\u001b[0m, in \u001b[0;36m_patch_loop.<locals>._run_once\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 100\u001b[0m heappop(scheduled)\n\u001b[1;32m 102\u001b[0m timeout \u001b[38;5;241m=\u001b[39m (\n\u001b[1;32m 103\u001b[0m \u001b[38;5;241m0\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m ready \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_stopping\n\u001b[1;32m 104\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28mmin\u001b[39m(\u001b[38;5;28mmax\u001b[39m(\n\u001b[1;32m 105\u001b[0m scheduled[\u001b[38;5;241m0\u001b[39m]\u001b[38;5;241m.\u001b[39m_when \u001b[38;5;241m-\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtime(), \u001b[38;5;241m0\u001b[39m), \u001b[38;5;241m86400\u001b[39m) \u001b[38;5;28;01mif\u001b[39;00m scheduled\n\u001b[1;32m 106\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m)\n\u001b[0;32m--> 107\u001b[0m event_list \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_selector\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mselect\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtimeout\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 108\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_process_events(event_list)\n\u001b[1;32m 110\u001b[0m end_time \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtime() \u001b[38;5;241m+\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_clock_resolution\n",
+ "File \u001b[0;32m/opt/conda/lib/python3.9/selectors.py:469\u001b[0m, in \u001b[0;36mEpollSelector.select\u001b[0;34m(self, timeout)\u001b[0m\n\u001b[1;32m 467\u001b[0m ready \u001b[38;5;241m=\u001b[39m []\n\u001b[1;32m 468\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 469\u001b[0m fd_event_list \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_selector\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpoll\u001b[49m\u001b[43m(\u001b[49m\u001b[43mtimeout\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmax_ev\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m 470\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mInterruptedError\u001b[39;00m:\n\u001b[1;32m 471\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m ready\n",
+ "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
]
}
],
"source": [
- "from tutorials.utils.tutorial_utils import visualize_history\n",
- "visualize_history(runner.simulation._history, runner.scenario, bokeh_port=5006)"
+ "from nuplan.planning.script.run_nuboard import main as main_nuboard\n",
+ "\n",
+ "# Run nuBoard\n",
+ "main_nuboard(cfg)"
]
},
{
"cell_type": "code",
"execution_count": null,
- "id": "887a51e2",
- "metadata": {},
- "outputs": [],
- "source": []
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "e0a12035",
- "metadata": {},
- "outputs": [],
- "source": []
- },
- {
- "cell_type": "code",
- "execution_count": null,
- "id": "67b67b86",
+ "id": "d824a7e4",
"metadata": {},
"outputs": [],
"source": []
diff --git a/nuplan/planning/script/builders/BUILD b/nuplan/planning/script/builders/BUILD
index 71ddd4ffeb3f991485bdb0dd90ac3489c52db6f9..24ea63288df93cc8222bcd5b88dacfbe4591e1b8 100644
--- a/nuplan/planning/script/builders/BUILD
+++ b/nuplan/planning/script/builders/BUILD
@@ -30,6 +30,7 @@ py_library(
"//nuplan/planning/metrics:metric_engine",
"//nuplan/planning/metrics/evaluation_metrics/common:drivable_area_compliance",
"//nuplan/planning/metrics/evaluation_metrics/common:driving_direction_compliance",
+ "//nuplan/planning/metrics/evaluation_metrics/common:is_relavant_agent_occluded",
"//nuplan/planning/metrics/evaluation_metrics/common:ego_acceleration",
"//nuplan/planning/metrics/evaluation_metrics/common:ego_expert_l2_error",
"//nuplan/planning/metrics/evaluation_metrics/common:ego_expert_l2_error_with_yaw",
diff --git a/nuplan/planning/script/config/common/simulation_metric/common_metrics.yaml b/nuplan/planning/script/config/common/simulation_metric/common_metrics.yaml
index 3233bfbe1c6f300367b4af4cf1d46affb307c256..f8c210f871b280f66a7eb47a675902f74c5de421 100644
--- a/nuplan/planning/script/config/common/simulation_metric/common_metrics.yaml
+++ b/nuplan/planning/script/config/common/simulation_metric/common_metrics.yaml
@@ -3,3 +3,4 @@ defaults:
- ego_mean_speed_statistics
- ego_expert_l2_error_statistics
- ego_expert_l2_error_with_yaw_statistics
+ - is_relavant_agent_occluded_statistics
diff --git a/nuplan/planning/script/config/common/simulation_metric/simulation_closed_loop_nonreactive_agents.yaml b/nuplan/planning/script/config/common/simulation_metric/simulation_closed_loop_nonreactive_agents.yaml
index 6efbed9866d96a1a4c781f2dd2d0c2e6ed98fa08..cfae686983c18f6554f320ed71e82071a3cda8c8 100644
--- a/nuplan/planning/script/config/common/simulation_metric/simulation_closed_loop_nonreactive_agents.yaml
+++ b/nuplan/planning/script/config/common/simulation_metric/simulation_closed_loop_nonreactive_agents.yaml
@@ -1,5 +1,5 @@
defaults:
- # - common_metrics # Uncomment this for common information about the scenario as specified in the config
+ - common_metrics # Uncomment this for common information about the scenario as specified in the config
- low_level: # Low level metrics
- ego_lane_change_statistics
- ego_jerk_statistics
diff --git a/nuplan/planning/script/config/common/simulation_metric/simulation_closed_loop_reactive_agents.yaml b/nuplan/planning/script/config/common/simulation_metric/simulation_closed_loop_reactive_agents.yaml
index 6efbed9866d96a1a4c781f2dd2d0c2e6ed98fa08..cfae686983c18f6554f320ed71e82071a3cda8c8 100644
--- a/nuplan/planning/script/config/common/simulation_metric/simulation_closed_loop_reactive_agents.yaml
+++ b/nuplan/planning/script/config/common/simulation_metric/simulation_closed_loop_reactive_agents.yaml
@@ -1,5 +1,5 @@
defaults:
- # - common_metrics # Uncomment this for common information about the scenario as specified in the config
+ - common_metrics # Uncomment this for common information about the scenario as specified in the config
- low_level: # Low level metrics
- ego_lane_change_statistics
- ego_jerk_statistics
diff --git a/nuplan/planning/script/config/common/simulation_metric/simulation_open_loop_boxes.yaml b/nuplan/planning/script/config/common/simulation_metric/simulation_open_loop_boxes.yaml
index a879f60601a9993e23d2eb82ce6e11dd4455940a..1801f38d855c8cee1e31b833dbae6d9421355370 100644
--- a/nuplan/planning/script/config/common/simulation_metric/simulation_open_loop_boxes.yaml
+++ b/nuplan/planning/script/config/common/simulation_metric/simulation_open_loop_boxes.yaml
@@ -1,5 +1,5 @@
defaults:
- # - common_metrics # Uncomment this for common information about the scenario as specified in the config
+ - common_metrics # Uncomment this for common information about the scenario as specified in the config
- low_level: # Low level metrics
- planner_expert_average_l2_error_within_bound_statistics
- high_level: # High level metrics that depend on low level metrics, they can also rely on the previously called high level metrics
diff --git a/nuplan/planning/simulation/history/simulation_history.py b/nuplan/planning/simulation/history/simulation_history.py
index d2b7042de5b7e0085303990ae03ab4005f3160f8..9ddbc83719f6cf2b3e3da21df43b8b1426ea074d 100644
--- a/nuplan/planning/simulation/history/simulation_history.py
+++ b/nuplan/planning/simulation/history/simulation_history.py
@@ -41,7 +41,7 @@ class SimulationHistory:
"""
self.map_api: AbstractMap = map_api
self.mission_goal = mission_goal
- # NOTE: This is just for visualization code, not used during simulation
+ # NOTE: This is used for visualization and for the is_relavant_agent_occluded metric
self.occlusion_masks = None
self.data: List[SimulationHistorySample] = data if data is not None else list()
@@ -80,6 +80,14 @@ class SimulationHistory:
:return An List of ego_states.
"""
return [sample.ego_state for sample in self.data]
+
+ @property
+ def extract_observations(self) -> List[EgoState]:
+ """
+ Extract observations in simulation history.
+ :return An List of observations.
+ """
+ return [sample.observation for sample in self.data]
@property
def interval_seconds(self) -> float:
@@ -92,4 +100,4 @@ class SimulationHistory:
elif len(self.data) < 2:
raise ValueError("Can't calculate the interval of a single-iteration simulation.")
- return float(self.data[1].iteration.time_s - self.data[0].iteration.time_s) # float cast is for mypy
+ return float(self.data[1].iteration.time_s - self.data[0].iteration.time_s) # float cast is for mypy
\ No newline at end of file
diff --git a/nuplan/planning/simulation/occlusion/wedge_occlusion_manager.py b/nuplan/planning/simulation/occlusion/wedge_occlusion_manager.py
index 366f33b4de312a1ab8e9b1b25981c42775146c3a..eb9cc8d1df40832b7c4258c075fc0fdfb110bd4e 100644
--- a/nuplan/planning/simulation/occlusion/wedge_occlusion_manager.py
+++ b/nuplan/planning/simulation/occlusion/wedge_occlusion_manager.py
@@ -41,8 +41,8 @@ class WedgeOcclusionManager(AbstractOcclusionManager):
# wedge based occlusion implementation. about half as fast and the occlusions flicker more but it should scale better if you have tons of occluders
def _determine_occlusions(self, observer: AgentState, targets:List[AgentState]) -> set:
- start = time.time()
- rads = np.linspace(0,2*math.pi,self.num_wedges+1)
+ #start = time.time()
+ rads = np.linspace(0, 2 * math.pi,self.num_wedges + 1)
wedges = dict()
for i in range(len(rads)-1):
@@ -121,6 +121,5 @@ class WedgeOcclusionManager(AbstractOcclusionManager):
for key in to_remove:
del wedges[key]
-
- print('elapsed time:', time.time() - start)
+ #print('elapsed time:', time.time() - start)
return not_occluded
\ No newline at end of file
diff --git a/tutorials/nuplan_framework.ipynb b/tutorials/nuplan_framework.ipynb
index 9ee1fe85a9fe509893aec5f3eb7f464c14380d6c..4b1e0a78b6215b7638b8d29441ec03e54d816927 100644
--- a/tutorials/nuplan_framework.ipynb
+++ b/tutorials/nuplan_framework.ipynb
@@ -1960,9 +1960,9 @@
"hash": "5aea00fa936a67f14323fa2d54163d7dc1f328c617e433b03a680f73ee2dd426"
},
"kernelspec": {
- "display_name": "nuplan",
+ "display_name": "Python 3 (ipykernel)",
"language": "python",
- "name": "nuplan"
+ "name": "python3"
},
"language_info": {
"codemirror_mode": {
@@ -1974,7 +1974,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.7"
+ "version": "3.9.16"
}
},
"nbformat": 4,
diff --git a/tutorials/nuplan_scenario_visualization.ipynb b/tutorials/nuplan_scenario_visualization.ipynb
index 837e1390efc6dac954a9caeb43fb982be1af7726..ac24fe6f5092dfbd1bebd8f23081bcb16c012b57 100644
--- a/tutorials/nuplan_scenario_visualization.ipynb
+++ b/tutorials/nuplan_scenario_visualization.ipynb
@@ -420,9 +420,9 @@
],
"metadata": {
"kernelspec": {
- "display_name": "nuplan",
+ "display_name": "Python 3 (ipykernel)",
"language": "python",
- "name": "nuplan"
+ "name": "python3"
},
"language_info": {
"codemirror_mode": {