Use unique MuJoCo XML patches in release wheels

Author: Trinkle23897

envpool/mujoco/BUILD

@@ -23,8 +23,8 @@ package(default_visibility = ["//visibility:public"])
 copy_to_directory(
     name = "gen_mujoco_gym_xml",
     srcs = [
-        "@mujoco_gym_xml",
         "//third_party/mujoco_gym_xml_patches:patches",
+        "@mujoco_gym_xml",
     ],
     out = "assets_gym",
     strip_prefix = "assets/",

envpool/mujoco/gym/hopper.h

@@ -77,7 +77,8 @@ class HopperEnv : public Env<HopperEnvSpec>, public MujocoEnv {
  public:
   HopperEnv(const Spec& spec, int env_id)
       : Env<HopperEnvSpec>(spec, env_id),
-        MujocoEnv(spec.config["base_path"_] + "/mujoco/assets_gym/hopper.xml",
+        MujocoEnv(spec.config["base_path"_] +
+                      "/mujoco/assets_gym/hopper_envpool.xml",
                   spec.config["frame_skip"_], spec.config["post_constraint"_],
                   spec.config["max_episode_steps"_]),
         terminate_when_unhealthy_(spec.config["terminate_when_unhealthy"_]),

envpool/mujoco/gym/registration.py

@@ -99,7 +99,7 @@
   if task == "Walker2d" and version == "v5":
     extra_args.update(
       {
-        "xml_file": "walker2d_v5.xml",
+        "xml_file": "walker2d_v5_envpool.xml",
         "legacy_healthy_reward": False,
       }
     )

envpool/mujoco/gym/walker2d.h

@@ -36,7 +36,7 @@ class Walker2dEnvFns {
         "ctrl_cost_weight"_.Bind(0.001), "terminate_when_unhealthy"_.Bind(true),
         "exclude_current_positions_from_observation"_.Bind(true),
         "legacy_healthy_reward"_.Bind(true),
-        "xml_file"_.Bind(std::string("walker2d.xml")),
+        "xml_file"_.Bind(std::string("walker2d_envpool.xml")),
         "forward_reward_weight"_.Bind(1.0), "healthy_reward"_.Bind(1.0),
         "healthy_z_min"_.Bind(0.8), "healthy_z_max"_.Bind(2.0),
         "healthy_angle_min"_.Bind(-1.0), "healthy_angle_max"_.Bind(1.0),