added more params for webapp | removed accidental merge conflict relicts

arena_bringup/launch/start_arena.launch

@@ -7,11 +7,21 @@
   <!-- FOR WEBAPP  -->
   <arg name="is_webapp_docker" default="false" /> 
   <arg name="task_id" default="" />
+  <arg name="app_token" default="" />
+  <arg name="app_token_key" default="" />
+  <arg name="task_finished_url" default="" />
+
   <param name="task_id" value="$(arg task_id)" />
+  <param name="app_token" value="$(arg app_token)" />
+  <param name="app_token_key" value="$(arg app_token_key)" />
+  <param name="task_finished_url" value="$(arg task_finished_url)" />
 
-  <node name="finished_task_publisher" type="finish_task_publisher" pkg="arena-utils" if="$(eval arg('is_webapp_docker') == true)" />
+  <node name="task_progress_publisher" type="task_progress_publisher.py" pkg="task_progress_publisher" if="$(eval arg('is_webapp_docker') == true)" />
   <!-- -->
 
+  <arg name="desired_resets" default="2" />
+  <param name="desired_resets" value="$(arg desired_resets)" />
+
 
   <!-- ___________ ARGS ___________ -->
   <!-- You can launch a single robot and his local_planner with arguments -->

arena_bringup/launch/start_training.launch

@@ -3,12 +3,18 @@
   <!-- FOR WEBAPP  -->
   <arg name="is_webapp_docker" default="false" /> 
   <arg name="task_id" default="" />
+  <arg name="app_token" default="" />
+  <arg name="app_token_key" default="" />
+  <arg name="task_finished_url" default="" />
+
   <param name="task_id" value="$(arg task_id)" />
+  <param name="app_token" value="$(arg app_token)" />
+  <param name="app_token_key" value="$(arg app_token_key)" />
+  <param name="task_finished_url" value="$(arg task_finished_url)" />
 
-  <node name="finished_task_publisher" type="finish_task_publisher" pkg="arena-utils" if="$(eval arg('is_webapp_docker') == true)" />
+  <node name="task_progress_publisher" type="task_progress_publisher.py" pkg="arena-utils" if="$(eval arg('is_webapp_docker') == true)" />
   <!-- -->
 
-
   <arg name="model" default="burger"      doc="model type [burger, jackal, ridgeback, agvota, rto, rto_real]"/>
 
   <!-- Here are the argument that may be frequently changed -->

training/scripts/train_agent.py

@@ -1,22 +1,8 @@
 #!/usr/bin/env python
-<<<<<<< HEAD
 import sys, rospy, time
-=======
-from typing import Type, Union
-
-import os, sys, rospy, time
 
 from std_msgs.msg import Empty
 
-from stable_baselines3 import PPO
-from stable_baselines3.common.vec_env import SubprocVecEnv, DummyVecEnv
-from stable_baselines3.common.callbacks import (
-    EvalCallback,
-    StopTrainingOnRewardThreshold,
-)
-from stable_baselines3.common.policies import ActorCriticPolicy, BasePolicy
->>>>>>> dev
-
 from rosnav.model.agent_factory import AgentFactory
 from tools.argsparser import parse_training_args
 from tools.general import *
@@ -61,6 +47,8 @@ def main():
     eval_cb = init_callbacks(config, train_env, eval_env, PATHS)
     model = get_ppo_instance(config, train_env, PATHS, AgentFactory)
 
+    rospy.on_shutdown(model.env.close())
+
     # start training
     start = time.time()
     try:
@@ -72,30 +60,17 @@ def main():
     except KeyboardInterrupt:
         print("KeyboardInterrupt..")
 
-<<<<<<< HEAD
-    rospy.on_shutdown(model.env.close())
-    print(f"Time passed: {time.time()-start}s. \n Training script will be terminated..")
-=======
-    model.learn(
-        total_timesteps=n_timesteps,
-        callback=eval_cb,
-        reset_num_timesteps=True,
-    )
 
-    # update the timesteps the model has trained in total
-    # update_total_timesteps_json(n_timesteps, PATHS)
+    print(f"Time passed: {time.time()-start}s. \n Training script will be terminated..")
 
     model.env.close()
-    print(f"Time passed: {time.time()-start}s")
-    print("Training script will be terminated")
 
     publisher = rospy.Publisher("training_finished", Empty, queue_size=10)
 
     for _ in range(10):
         publisher.publish(Empty())
         rospy.sleep(0.1)
 
->>>>>>> dev
     sys.exit()