[0.2.3][Hydra Migration] Made agents to be of type Dict[str, AgentConfig] and added migration readme #1007

rpartsey · 2022-11-19T17:39:41Z

Motivation and Context

Made habitat.simulator.agents to be of type Dict[str, AgentConfig] to allow the possibility to dynamically configure the number of agents (in the future will simplify the multi-agent support).
Added migration README

How Has This Been Tested

Types of changes

Docs change / refactoring / dependency upgrade
Bug fix (non-breaking change which fixes an issue)
New feature (non-breaking change which adds functionality)
Breaking change (fix or feature that would cause existing functionality to change)

Checklist

My code follows the code style of this project.
My change requires a change to the documentation.
I have updated the documentation accordingly.
I have read the CONTRIBUTING document.
I have completed my CLA (see CONTRIBUTING)
I have added tests to cover my changes.
All new and existing tests passed.

vincentpierre · 2022-11-21T18:43:41Z

habitat-lab/habitat/config/default_structured_configs.py

+    # agent_0: AgentConfig = AgentConfig()
+    # agent_1: AgentConfig = (
+    #     MISSING  # temporary multi-agent rearrangement config fix
+    # )


Suggested change

# agent_0: AgentConfig = AgentConfig()

# agent_1: AgentConfig = (

# MISSING # temporary multi-agent rearrangement config fix

# )

vincentpierre · 2022-11-21T18:50:23Z

habitat-lab/habitat/sims/habitat_simulator/habitat_simulator.py

@@ -306,7 +308,7 @@ def create_sim_config(
        sim_config.scene_id = self.habitat_config.scene
        agent_config = habitat_sim.AgentConfiguration()
        overwrite_config(
-            config_from=self._get_agent_config(),
+            config_from=self._get_default_agent_config(),


This is not really default. Should it be _get_any_agent_config?

vincentpierre · 2022-11-21T18:51:03Z

habitat-lab/habitat/tasks/nav/nav.py

@@ -75,8 +75,7 @@ def merge_sim_episode_config(
        episode.start_position is not None
        and episode.start_rotation is not None
    ):
-        agent_name = sim_config.agents[sim_config.default_agent_id]
-        agent_cfg = getattr(sim_config, agent_name)
+        agent_cfg = next(iter(sim_config.agents.values()))


I think there should be an assert that len(sim_config.agents.values()) == 1

vincentpierre · 2022-11-21T18:51:50Z

habitat-lab/habitat/utils/gym_definitions.py

@@ -65,12 +65,15 @@ def _make_habitat_gym_env(
    config = habitat.get_config(cfg_file_path)
    if use_render_mode:
        with habitat.config.read_write(config):
+            default_agent = next(


assert len(config.habitat.simulator.agents.values()) == 1

vincentpierre · 2022-11-21T19:12:14Z

habitat-lab/habitat/config/default_structured_configs.py

@@ -1177,85 +1179,85 @@ class HabitatConfig(HabitatBaseConfig):

 # Simulator Sensors
 cs.store(
-    package="habitat.simulator.agent_0.sim_sensors.rgb_sensor",
+    package="habitat.simulator.agents.agent_0.sim_sensors.rgb_sensor",


Kind of awkward to have agent_0 in here now...

…configuration updates; updated docs

rpartsey · 2022-11-27T09:30:52Z

habitat-baselines/habitat_baselines/config/default.py

-        )
-
-    OmegaConf.set_readonly(cfg, True)
+    cfg = _habitat_get_config(config_paths, overrides, configs_dir)


Habitat-Lab's configs are included in the Baselines config by specifying corresponding Habitat-Lab input config in the defaults list. In other words the Baselines config in an extension of the Habitat-Lab config. Therefore, Lab and Baselines configs have to be read/created by the same function. To make sure that all the possible custom validations/configs updates via code in Lab's config are also applied in the Baselines' get_config.

Thank you for deduplicating the code.

* added habitat-lab config readme

vincentpierre · 2022-11-28T20:00:54Z

habitat-lab/habitat/config/CONFIG.md

@@ -0,0 +1,482 @@
+Habitat-Lab Configuration System


Rename to README.md

vincentpierre · 2022-11-28T20:01:05Z

habitat-lab/habitat/config/CONFIG.md

+
+@dataclass
+class CustomStructuredConfig:
+    custom_config_key: KeyType = DefayltValue


Suggested change

custom_config_key: KeyType = DefayltValue

custom_config_key: KeyType = DefaultValue

vincentpierre · 2022-11-28T22:52:40Z

habitat-lab/habitat/config/CONFIG.md

+## Hydra Concepts used in Habitat-Lab
+
+With Hydra, the [Output Config](https://hydra.cc/docs/advanced/terminology/#output-config)
+is composed dynamically at the run time from


Suggested change

is composed dynamically at the run time from

is composed dynamically at run time from

vincentpierre · 2022-11-28T22:53:37Z

habitat-lab/habitat/config/CONFIG.md

+config validation schemas to ensure that all the required fields are set and match the required type, and also as a
+configs, in place of configuration yaml files. All Structured Configs are registered in
+the [ConfigStore](https://hydra.cc/docs/tutorials/structured_config/config_store/) - in-memory
+Structured Configs registry. Habitat-Lab's Structured Configs are defined and registered to the ConfigStore in the


Suggested change

Structured Configs registry. Habitat-Lab's Structured Configs are defined and registered to the ConfigStore in the

Structured Configs registry. Habitat-Lab's Structured Configs are defined and registered to the ConfigStore in

vincentpierre · 2022-11-28T22:53:55Z

habitat-lab/habitat/config/CONFIG.md

+Structured Configs registry. Habitat-Lab's Structured Configs are defined and registered to the ConfigStore in the
+[habitat-lab/habitat/config/default_structured_configs.py](default_structured_configs.py).
+
+Similar configs are grouped in the [Config Groups](https://hydra.cc/docs/advanced/terminology/#config-group)


Suggested change

Similar configs are grouped in the [Config Groups](https://hydra.cc/docs/advanced/terminology/#config-group)

Similar configs are grouped in [Config Groups](https://hydra.cc/docs/advanced/terminology/#config-group)

vincentpierre · 2022-11-28T22:58:04Z

habitat-lab/habitat/config/CONFIG.md

+For example, Embodied AI task specifications supported by Habitat-Lab are placed in the `habitat/task` Config Group
+and to use the PointNav task just add `habitat/task: pointnav` line to your config's Defaults List.
+
+Where the content of each Input Config (Config Group Option) is placed in the Output Config is determined via


Suggested change

Where the content of each Input Config (Config Group Option) is placed in the Output Config is determined via

Each Input Config's position in the Output Config is determined via

vincentpierre · 2022-11-28T23:01:19Z

habitat-lab/habitat/config/CONFIG.md

+
+## What's changed?
+<details>
+<summary>Expand to see the Habitat-Lab input PointNav benchmark config example.</summary>


Suggested change

<summary>Expand to see the Habitat-Lab input PointNav benchmark config example.</summary>

PointNav benchmark: Input config.</summary>

vincentpierre · 2022-11-28T23:02:32Z

habitat-lab/habitat/config/CONFIG.md

+</details>
+
+<details>
+<summary>Expand to see the corresponding composed output config.</summary>


Suggested change

<summary>Expand to see the corresponding composed output config.</summary>

<summary>

PointNav benchmark: Output config.</summary>

vincentpierre · 2022-11-28T23:07:29Z

habitat-lab/habitat/config/CONFIG.md

+- `agent.sensors` is renamed to `agent.sim_sensors`,  `task.sensors` is renamed to `task.lab_sensors`.
+- Actions, agents, measurements and sensors configs are not directly attached to the simulator, task or agent
+  config nodes but are added to `task.actions`, `simulator.agents`, `task.measurements`, `task.lab_sensors`,
+  `simulator.agent_name.sim_sensors` sub-nodes of type `Dict[str, CorrespondingConfigNodeType]`


It needs to be clear that agent name is not hard coded string.

Suggested change

`simulator.agent_name.sim_sensors` sub-nodes of type `Dict[str, CorrespondingConfigNodeType]`

`simulator.<agent_name>.sim_sensors` sub-nodes of type `Dict[str, CorrespondingConfigNodeType]`

vincentpierre · 2022-11-29T00:08:02Z

habitat-lab/habitat/config/benchmark/rearrange/pick.yaml

+#          head_rgb_sensor:
+#              width: 256
+#              height: 256
+#          arm_depth_sensor:
+#            height: 128
+#            width: 128
+#          arm_rgb_sensor:
+#            height: 128
+#            width: 128


Suggested change

# head_rgb_sensor:

# width: 256

# height: 256

# arm_depth_sensor:

# height: 128

# width: 128

# arm_rgb_sensor:

# height: 128

# width: 128

vincentpierre · 2022-11-29T00:09:23Z

habitat-lab/habitat/config/default.py

+
+    # Check if the `habitat.simulator.agents_order`
+    # is set and matches the agents' keys in `habitat.simulator.agents`.
+    assert set(sim_config.agents_order) == set(sim_config.agents.keys()), (


Assert on the length of the two as well. agents_order could have duplicates

mathfac · 2022-11-29T00:19:38Z

habitat-lab/habitat/config/default.py

-def get_full_config_path(
-    config_path: str, default_configs_dir: str = _HABITAT_CFG_DIR
-) -> str:
+def get_full_config_path(config_path: str, configs_dir: str) -> str:


Need string doc here to describe the function arguments and output.

mathfac · 2022-11-29T00:21:25Z

habitat-lab/habitat/config/default.py

    if osp.exists(proposed_full_path):
        return osp.abspath(proposed_full_path)

    raise RuntimeError(f"No file found for config '{config_path}'")


+get_full_habitat_config_path = partial(


Same here. This functions will be frequently used and should have better documentation.

habitat-lab/habitat/config/default.py

mathfac · 2022-11-29T00:25:59Z

habitat-lab/habitat/config/default.py

@@ -57,6 +73,19 @@ def get_config(
            overrides=overrides if overrides is not None else [],
        )

+    # In the single-agent setup use the agent's key from `habitat.simulator.agents`.
+    sim_config = cfg.habitat.simulator
+    if len(sim_config.agents) == 1:


Let's remove this non documented functionality. We plan to create "main_agent" for this purpose to be the default, those it won't be needed..

We agreed to create just an alias. Actually, I'd leave it as it is forcing users to explicitly define / chose the agent. @vincentpierre your thoughts?

I think we should leave it as is. I do not want to have main_agent be defined in code (and that includes the default_structured_config file) but only be defined in the yamls that we provide. Nothing should prevent the user to name their unique agent anything they want and not need to modify agent_order

I don't like the term "main" agent as "main" is an very overloaded term. Is there a better term we can use? Even agent_0 is more specific and easier to understand in my opinion.

mathfac · 2022-11-29T00:34:14Z

habitat-lab/habitat/config/default.py

+    # Check if the `habitat.simulator.agents_order`
+    # is set and matches the agents' keys in `habitat.simulator.agents`.
+    assert set(sim_config.agents_order) == set(sim_config.agents.keys()), (
+        "habitat.simulator.agents_order should be set explicitly "


Let's be more explicit:

Suggested change

"habitat.simulator.agents_order should be set explicitly "

"habitat.simulator.agents_order({sim_config.agents_order}) should be set explicitly "

mathfac · 2022-11-29T00:34:41Z

habitat-lab/habitat/config/default.py

+    # is set and matches the agents' keys in `habitat.simulator.agents`.
+    assert set(sim_config.agents_order) == set(sim_config.agents.keys()), (
+        "habitat.simulator.agents_order should be set explicitly "
+        "and match the agents' keys in habitat.simulator.agents"


Suggested change

"and match the agents' keys in habitat.simulator.agents"

"and match the agents' keys in habitat.simulator.agents({set(sim_config.agents.keys())})"

mathfac · 2022-11-29T00:36:43Z

habitat-lab/habitat/config/default_structured_configs.py

+    # if the number of agents is greater than one,
+    # then agents_order has to be set explicitly too


Maybe something like this:

Suggested change

# if the number of agents is greater than one,

# then agents_order has to be set explicitly too

# Agents order that will be used by simulator step and trainer policy.

agent_order is only used to know in which order to return the observations and accept the actions when using the environment API correct?
In any case, we should explain here what the agent_order does for the user

Defining agent order like this seems a bit of an overkill. Why can't the order of the agents defined in the YAML just be the order of the agents? That seems a lot more intuitive to me instead of having this foreign key. Why cant the agents key be a list of of agents instead? Then the order is defined implicitly by the order the agents are added to the list. That's how it was going to be before in yacs with mutliagent support. Don't know why we changed the agent configs from an ordered list to a dict.

mathfac · 2022-11-29T06:34:57Z

habitat-lab/habitat/config/default.py

+)
+
+
+def get_agent_config(sim_config, agent_id: Optional[int] = None):


Added doc string and type for sim_config:

Suggested change

def get_agent_config(sim_config, agent_id: Optional[int] = None):

def get_agent_config(sim_config: DictConfig, agent_id: Optional[int] = None) -> DictConfig:

r"""Called to get agent's config node of default agent or based on index of the agent.

:param sim_config: config of :ref:`core.simulator.Simulator`.

:param agent_id: index of the agent (relevant for multi-agent setup).

:return: relevant agent's config.

"""

mathfac · 2022-11-29T06:39:08Z

habitat-lab/habitat/config/default.py

@@ -16,36 +17,51 @@
    HabitatConfigPlugin,
    register_hydra_plugin,
 )
+from habitat.config.read_write import read_write

 _HABITAT_CFG_DIR = osp.dirname(inspect.getabsfile(inspect.currentframe()))
 # This is equivalent to doing osp.dirname(osp.abspath(__file__))


Suggested change

# This is equivalent to doing osp.dirname(osp.abspath(__file__))

# Habitat config directory inside the installed package.

# Used to access default predefined configs.

# This is equivalent to doing osp.dirname(osp.abspath(__file__))

mathfac · 2022-11-29T06:41:05Z

habitat-lab/habitat/config/test/habitat_all_sensors_test.yaml

-          width: 256
-          height: 256
+    agents:
+      rgbd_agent:


As we discussed let's have main_agent node for all single agent usecases.

mathfac · 2022-11-29T06:41:32Z

habitat-lab/habitat/config/test/habitat_mp3d_eqa_test.yaml

-          position: [0, 1.09, 0]
-          type: HabitatSimRGBSensor
+    agents:
+      rgbds_agent:


Suggested change

rgbds_agent:

main_agent:

mathfac · 2022-11-29T06:42:43Z

docs/pages/habitat-lab-demo.rst

@@ -128,7 +129,8 @@ the config object.
    config = habitat.get_config(config_paths="benchmark/nav/pointnav/pointnav_mp3d.yaml")
    with read_write(config):
        config.habitat.dataset.split = "val"
-        config.habitat.simulator.agent_0.sim_sensors.update(
+        agent_config = get_agent_config(config.habitat.simulator)


In examples let's be more explicit:

Suggested change

agent_config = get_agent_config(config.habitat.simulator)

agent_config = get_agent_config(sim_config=config.habitat.simulator)

mathfac · 2022-11-29T06:43:14Z

docs/pages/view-transform-warp.rst

@@ -31,14 +32,15 @@ validate that transformation comparing projected and original views.
        config = habitat.get_config(config_paths="benchmark/nav/pointnav/pointnav_habitat_test.yaml")
        with read_write(config):
            config.habitat.dataset.split = "val"
-            config.habitat.simulator.agent_0.sim_sensors.depth_sensor.normalize_depth = False
+            agent_config = get_agent_config(config.habitat.simulator)


In examples let's be more explicit:

Suggested change

agent_config = get_agent_config(config.habitat.simulator)

agent_config = get_agent_config(sim_config=config.habitat.simulator)

mathfac · 2022-11-29T06:43:46Z

examples/interactive_play.py

@@ -593,7 +594,8 @@ def has_pygame():

        if not args.same_task:
            sim_config.debug_render = True
-            sim_config.agent_0.sim_sensors.update(
+            agent_config = get_agent_config(sim_config)


Suggested change

agent_config = get_agent_config(sim_config)

agent_config = get_agent_config(sim_config=sim_config)

mathfac · 2022-11-29T06:44:07Z

examples/tutorials/nb_python/Habitat2_Quickstart.py

@@ -92,7 +93,8 @@ def insert_render_options(config):
    # Added settings to make rendering higher resolution for better visualization
    with habitat.config.read_write(config):
        config.habitat.simulator.concur_render = False
-        config.habitat.simulator.agent_0.sim_sensors.update(
+        agent_config = get_agent_config(config.habitat.simulator)


Suggested change

agent_config = get_agent_config(config.habitat.simulator)

agent_config = get_agent_config(sim_config=config.habitat.simulator)

mathfac · 2022-11-29T06:44:23Z

examples/tutorials/nb_python/Habitat2_Quickstart.py

-      robot_urdf: ./data/robots/hab_fetch/robots/hab_fetch.urdf
-      robot_type: "FetchRobot"
+    agents:
+      agent_0:


Suggested change

agent_0:

main_agent:

habitat-baselines/habitat_baselines/config/default.py

mathfac · 2022-11-29T06:45:57Z

habitat-baselines/habitat_baselines/config/default.py

-        )
-
-    OmegaConf.set_readonly(cfg, True)
+    cfg = _habitat_get_config(config_paths, overrides, configs_dir)


Thank you for deduplicating the code.

habitat-baselines/habitat_baselines/config/default.py

habitat-baselines/habitat_baselines/config/eqa/il_pacman_nav.yaml

mathfac · 2022-11-29T06:56:58Z

habitat-lab/habitat/config/CONFIG.md

+    # Other SimulatorConfig keys are omitted in this code snippet
+    seed: int = II("habitat.seed")
+```
+- [Multirun](https://hydra.cc/docs/tutorials/basic/running_your_app/multi-run/). For example, launching 3 experiments with three different learning rates:


Suggested change

- [Multirun](https://hydra.cc/docs/tutorials/basic/running_your_app/multi-run/). For example, launching 3 experiments with three different learning rates:

- [Parameter sweeping and multirun](https://hydra.cc/docs/tutorials/basic/running_your_app/multi-run/). For example, launching 3 experiments with three different learning rates:

mathfac · 2022-11-29T09:23:57Z

habitat-lab/habitat/config/CONFIG.md

+python -u habitat_baselines/run.py --exp-config config.yaml --run-type train –-multirun \
+habitat_baselines.rl.ppo.lr 2.5e-4,2.5e-5,2.5e-6
+```
+


Suggested change

- Seamless [SLURM](https://slurm.schedmd.com/documentation.html) integration through [Submitit Launcher](https://hydra.cc/docs/plugins/submitit_launcher/). To enable the feature Submitit plugin should be installed: `pip install hydra-submitit-launcher --upgrade` and specify launcher `hydra/launcher=submitit_slurm` in the command line:

```bash

python -u habitat_baselines/run.py --exp-config config.yaml --run-type train –-multirun \

habitat_baselines.rl.ppo.lr 2.5e-4,2.5e-5,2.5e-6 hydra/launcher=submitit_slurm

mathfac · 2022-11-29T09:24:26Z

habitat-lab/habitat/config/CONFIG.md

+  - /habitat/simulator/agents@habitat.simulator.agents.agent_0: depth_head_agent
+  - /habitat/simulator/agents@habitat.simulator.agents.agent_1: depth_head_agent
+```
+- Making the config key required by setting its value to `MISSING`. For example, we require the user to explicitly


Let's move this point as the last one.

mathfac · 2022-11-29T09:25:46Z

habitat-lab/habitat/config/CONFIG.md

+  (not all possible actions, agents, measurements and sensors config nodes).
+- `task.possible_actions` is removed.
+
+### New features


That would be good to mention that functionality is the result of the migration:

Suggested change

### New features

### New functionality enabled

mathfac · 2022-11-29T09:33:40Z

test/test_baselines_hydra.py

                # "+habitat=benchmark/rearrange/pick",
-                # "+sim_sensors@habitat.simulator.agent_0.sim_sensors.arm_rgb_sensor=arm_rgb_sensor",
+                # "+sim_sensors@habitat.simulator.agents.agent_0.sim_sensors.arm_rgb_sensor=arm_rgb_sensor",


We should add the comments why the lines are commented out or remove them.

mathfac · 2022-11-29T09:37:02Z

habitat-lab/habitat/tasks/nav/nav.py

@@ -75,8 +75,10 @@ def merge_sim_episode_config(
        episode.start_position is not None
        and episode.start_rotation is not None
    ):
-        agent_name = sim_config.agents[sim_config.default_agent_id]
-        agent_cfg = getattr(sim_config, agent_name)
+        default_agent_name = sim_config.agents_order[


We should use get_agent_config method here to be consistent.

mathfac · 2022-11-29T09:39:08Z

habitat-lab/habitat/sims/habitat_simulator/habitat_simulator.py

+            len(
+                self.sim_config.agents[
+                    self.habitat_config.default_agent_id
+                ].action_space


Use get_agent_config for consistency:

Suggested change

len(

self.sim_config.agents[

self.habitat_config.default_agent_id

].action_space

len(

get_agent_config(self.sim_config).action_space

Here self.sim_config is not of type DictConfig it is of type habitat_sim.Configuration. So the code should be kept as it is.

vincentpierre · 2022-11-29T17:34:06Z

habitat-lab/habitat/config/default.py

@@ -57,6 +73,19 @@ def get_config(
            overrides=overrides if overrides is not None else [],
        )

+    # In the single-agent setup use the agent's key from `habitat.simulator.agents`.
+    sim_config = cfg.habitat.simulator
+    if len(sim_config.agents) == 1:


I think we should leave it as is. I do not want to have main_agent be defined in code (and that includes the default_structured_config file) but only be defined in the yamls that we provide. Nothing should prevent the user to name their unique agent anything they want and not need to modify agent_order

habitat-lab/habitat/config/default.py

vincentpierre · 2022-11-29T17:36:42Z

habitat-lab/habitat/config/default_structured_configs.py

+    # if the number of agents is greater than one,
+    # then agents_order has to be set explicitly too


agent_order is only used to know in which order to return the observations and accept the actions when using the environment API correct?
In any case, we should explain here what the agent_order does for the user

vincentpierre · 2022-11-29T17:37:20Z

habitat-lab/habitat/config/habitat/simulator/agents/depth_agent.yaml

@@ -0,0 +1,5 @@
+# @package habitat.simulator.agents.depth_agent


TODO later (not in this PR)
Make a 256 version of these to reduce complexity of benchmark configs

Skylion007 · 2022-11-29T19:25:37Z

habitat-lab/habitat/config/benchmark/nav/imagenav/imagenav_base.yaml

-          width: 256
-          height: 256
+    agents:
+      main_agent:


Why are we changing agent_0 to main_agent? Seems less intuitive to me, especially for multi_agent support.

Skylion007 · 2022-11-29T19:33:02Z

habitat-lab/habitat/config/default_structured_configs.py

-        MISSING  # temporary multi-agent rearrangement config fix
-    )
+    # we assume agent(s) to be set explicitly
+    agents: Dict[str, AgentConfig] = MISSING


Suggested change

agents: Dict[str, AgentConfig] = MISSING

agents: List[AgentConfig] = MISSING

Why not that? Then agent order is defined, each agent has a unique ID given by its order in the list, right? Or is there some reason we can't do that. In the event taht does not work, does Hydrasupport that Dict just being an OrderedDict instead? ANd then the order still matters?
Or even more cleverly:

Suggested change

agents: Dict[str, AgentConfig] = MISSING

agents: OrderedDict[str, AgentConfig] = MISSING

Skylion007

I actually do not like how agent_order now needs to manually specified. It is error prone and that information can be specified by the user in a more intuitive way. We already have an agent order implicitly from the order the agents are defined in the YAML, can we need use that, either via an OrderedDict or via List?

Skylion007 · 2022-11-29T19:40:18Z

examples/tutorials/nb_python/Habitat2_Quickstart.py

-      robot_urdf: ./data/robots/hab_fetch/robots/hab_fetch.urdf
-      robot_type: "FetchRobot"
+    agents:
+      main_agent:


Okay, so why are we naming the agent with dictionary keys instead of just introducing a harmless name object to the AgentConfig? Seems like we are just complicating multiagent support. Is there a good reason to do this? Is it due to limitations in YAML? That dictionaries are easier to merge together? Can we preserve the ordering of the dictionary somehow to specify agent order?

Better yet can we copy the way defaults is defined at the top fo the yaml

agents: - main_agent: ... - agent_1: ...

Skylion007

Fine without multiagent issues.

Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com>

…/habitat-lab into hydra_migration_3_v2_agents_upd

* updated md files * updated habitat2_gym_tutorial * updated Habitat2_Quickstart * updated Habitat_Lab * updated docs * partially removed YACS (note there are some inpedendent parts of code that still rely on YACS) * made all DictConfig imports to be from omegaconf using if TYPE_CHECKING * removed link to the outdated notebook * fixed tests * [Hydra Migration] Removing yacs and migratting the dataset generation code of rearrange (#1001) * removing yacs and migratting the dataset generation code of rearrange * addressing comments * typechecking corrections Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * renaming SensorConfig to LabSensorConfig * fixing svg link * fixing svg; * using relative path is better for svg * addressed comments * [0.2.3][Hydra Migration] Made agents to be of type Dict[str, AgentConfig] and added migration readme (#1007) * made habitat.simulator.agents be a Dict[str, AgentConfig] * addressed comments * refactored sim agents configuration * moved rearrange agents declarations to agents folder; finised agents configuration updates; updated docs * updated get_config import * Cfg sys migration readme (#2) * added habitat-lab config readme * addressed most of the comments * added main_agent alias * renamed habitat config reagme and added baselines config readme * Update habitat-lab/habitat/config/default.py Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * added desctiption for the agents_order field * added a dot :) Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * fixing conflicts Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com>

* [v0.2.3] Add robot document (#1018) * [v0.2.3] [Stretch] play stretch demo (#1014) * [Fix] Documentation gym example path (#1016) Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * [0.2.3]Hydra migration 3 v2 (#1000) * updated md files * updated habitat2_gym_tutorial * updated Habitat2_Quickstart * updated Habitat_Lab * updated docs * partially removed YACS (note there are some inpedendent parts of code that still rely on YACS) * made all DictConfig imports to be from omegaconf using if TYPE_CHECKING * removed link to the outdated notebook * fixed tests * [Hydra Migration] Removing yacs and migratting the dataset generation code of rearrange (#1001) * removing yacs and migratting the dataset generation code of rearrange * addressing comments * typechecking corrections Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * renaming SensorConfig to LabSensorConfig * fixing svg link * fixing svg; * using relative path is better for svg * addressed comments * [0.2.3][Hydra Migration] Made agents to be of type Dict[str, AgentConfig] and added migration readme (#1007) * made habitat.simulator.agents be a Dict[str, AgentConfig] * addressed comments * refactored sim agents configuration * moved rearrange agents declarations to agents folder; finised agents configuration updates; updated docs * updated get_config import * Cfg sys migration readme (#2) * added habitat-lab config readme * addressed most of the comments * added main_agent alias * renamed habitat config reagme and added baselines config readme * Update habitat-lab/habitat/config/default.py Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * added desctiption for the agents_order field * added a dot :) Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * fixing conflicts Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * Fix links in MANIFEST.in (#991) * Fix broken links in MANIFEST.in * fixig video_render_view * fixing issue with gym maker * editing test that changed with recent config modifications * [v0.2.3] Fixing mypy (#1023) * Unpin habitat-lab requirements * Reorder dependencies * Fix mypy config * remove unnecessary mypy args * Update mypy version * Fix python lint * Fix pre-commit mypy further * fix mypy ignored errors * fixing mypy * fixing mypy * more accurate type:ignores * adding docs/ to exclude mypy * trying to fix python_lint * exlude arguments * exlude arguments * Container needs to be a string Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * removing the gym envs keys that were removed Co-authored-by: Jimmy Yang <55121504+jimmytyyang@users.noreply.github.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> Co-authored-by: Ruslan <partsey2412@gmail.com> Co-authored-by: Mikaël Dallaire Côté <110583667+0mdc@users.noreply.github.com> Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com>

* [bugfix] Reconfigure the sim even if the sim is not None * addressing comments, better way to decide wether or not to reconfigure sim_obj * same fix for the robot base * Added back caching to pick task * Fixed up nav to obj task * Fixed up open and close skills * Added back inits for nav to obj * Batched HRL policy * Updated task conditions * Fixed key mismatch in TP-SRL * Fixed issue with nav to obj skill * Fixed early stop issue * Fixed bad filter condition * Updated training configs * Refactored and cleaned up config files * Fixed up reward. Renamed configs * Added wandb resume * Refactored eval with debug sensor code. * Added get policy info * Hopefully fixed resuming, added safe drop * Added improved Fetch model * Removed refactored files * Added back default * Removed habitat folder * Fixed pick task * Fixed navigation skill * Removed redundancy in skill training configs * Addressed some comments * Removed nav pick * Update .gitignore Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * PR comments * [v0.2.3] Hab2 fixup : merging main and resolving conflicts (#1022) * [v0.2.3] Add robot document (#1018) * [v0.2.3] [Stretch] play stretch demo (#1014) * [Fix] Documentation gym example path (#1016) Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * [0.2.3]Hydra migration 3 v2 (#1000) * updated md files * updated habitat2_gym_tutorial * updated Habitat2_Quickstart * updated Habitat_Lab * updated docs * partially removed YACS (note there are some inpedendent parts of code that still rely on YACS) * made all DictConfig imports to be from omegaconf using if TYPE_CHECKING * removed link to the outdated notebook * fixed tests * [Hydra Migration] Removing yacs and migratting the dataset generation code of rearrange (#1001) * removing yacs and migratting the dataset generation code of rearrange * addressing comments * typechecking corrections Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * renaming SensorConfig to LabSensorConfig * fixing svg link * fixing svg; * using relative path is better for svg * addressed comments * [0.2.3][Hydra Migration] Made agents to be of type Dict[str, AgentConfig] and added migration readme (#1007) * made habitat.simulator.agents be a Dict[str, AgentConfig] * addressed comments * refactored sim agents configuration * moved rearrange agents declarations to agents folder; finised agents configuration updates; updated docs * updated get_config import * Cfg sys migration readme (#2) * added habitat-lab config readme * addressed most of the comments * added main_agent alias * renamed habitat config reagme and added baselines config readme * Update habitat-lab/habitat/config/default.py Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * added desctiption for the agents_order field * added a dot :) Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * fixing conflicts Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * Fix links in MANIFEST.in (#991) * Fix broken links in MANIFEST.in * fixig video_render_view * fixing issue with gym maker * editing test that changed with recent config modifications * [v0.2.3] Fixing mypy (#1023) * Unpin habitat-lab requirements * Reorder dependencies * Fix mypy config * remove unnecessary mypy args * Update mypy version * Fix python lint * Fix pre-commit mypy further * fix mypy ignored errors * fixing mypy * fixing mypy * more accurate type:ignores * adding docs/ to exclude mypy * trying to fix python_lint * exlude arguments * exlude arguments * Container needs to be a string Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * removing the gym envs keys that were removed Co-authored-by: Jimmy Yang <55121504+jimmytyyang@users.noreply.github.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> Co-authored-by: Ruslan <partsey2412@gmail.com> Co-authored-by: Mikaël Dallaire Côté <110583667+0mdc@users.noreply.github.com> Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com> * Merged * Removed bad file and fixed IK * Env fixes * Fixed precommit * Fixed eval and fixed wandb logging * Fix for the resume in VER and the force threshold for open tasks * Update habitat-baselines/habitat_baselines/rl/ddppo/policy/resnet_policy.py Co-authored-by: Aaron Gokaslan <skylion.aaron@gmail.com> * tp-srl fixes * tp-srl fixes * Fixed more tests * Cleaned up nav action * Added tests for all tasks in baseline trainer * Corrected test overrides * Improved agents loop in resnet_policy Improved agents loop in resnet_policy Co-authored-by: Ruslan <partsey2412@gmail.com> * Fixed policy docs and improvements. Added docs to spawn robot method. * [0.2.3] 959 PR review suggested changes (#1029) * removed habitat-baselines/habitat_baselines/config path prefix * improved adding render view uuids to the gym obs_keys * made config keys lowercase * Updated docs * Update habitat-lab/habitat/tasks/rearrange/rearrange_task.py Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * Update habitat-baselines/habitat_baselines/rl/ddppo/policy/resnet_policy.py Co-authored-by: Aaron Gokaslan <skylion.aaron@gmail.com> * Fixed formatting * tp-srl config fix * fixing documentation ? Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> Co-authored-by: ASzot <me@andrewszot.com> Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> Co-authored-by: Jimmy Yang <55121504+jimmytyyang@users.noreply.github.com> Co-authored-by: Ruslan <partsey2412@gmail.com> Co-authored-by: Mikaël Dallaire Côté <110583667+0mdc@users.noreply.github.com> Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com> Co-authored-by: Aaron Gokaslan <skylion.aaron@gmail.com> Co-authored-by: Oleksandr <maksymets.o@gmail.com> Co-authored-by: Oleksandr Maksymets <maksymets@gmail.com> Co-authored-by: SergioArnaud <sercharnaud@gmail.com>

* updated md files * updated habitat2_gym_tutorial * updated Habitat2_Quickstart * updated Habitat_Lab * updated docs * partially removed YACS (note there are some inpedendent parts of code that still rely on YACS) * made all DictConfig imports to be from omegaconf using if TYPE_CHECKING * removed link to the outdated notebook * fixed tests * [Hydra Migration] Removing yacs and migratting the dataset generation code of rearrange (facebookresearch#1001) * removing yacs and migratting the dataset generation code of rearrange * addressing comments * typechecking corrections Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * renaming SensorConfig to LabSensorConfig * fixing svg link * fixing svg; * using relative path is better for svg * addressed comments * [0.2.3][Hydra Migration] Made agents to be of type Dict[str, AgentConfig] and added migration readme (facebookresearch#1007) * made habitat.simulator.agents be a Dict[str, AgentConfig] * addressed comments * refactored sim agents configuration * moved rearrange agents declarations to agents folder; finised agents configuration updates; updated docs * updated get_config import * Cfg sys migration readme (facebookresearch#2) * added habitat-lab config readme * addressed most of the comments * added main_agent alias * renamed habitat config reagme and added baselines config readme * Update habitat-lab/habitat/config/default.py Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * added desctiption for the agents_order field * added a dot :) Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * fixing conflicts Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com>

…okresearch#959) * [bugfix] Reconfigure the sim even if the sim is not None * addressing comments, better way to decide wether or not to reconfigure sim_obj * same fix for the robot base * Added back caching to pick task * Fixed up nav to obj task * Fixed up open and close skills * Added back inits for nav to obj * Batched HRL policy * Updated task conditions * Fixed key mismatch in TP-SRL * Fixed issue with nav to obj skill * Fixed early stop issue * Fixed bad filter condition * Updated training configs * Refactored and cleaned up config files * Fixed up reward. Renamed configs * Added wandb resume * Refactored eval with debug sensor code. * Added get policy info * Hopefully fixed resuming, added safe drop * Added improved Fetch model * Removed refactored files * Added back default * Removed habitat folder * Fixed pick task * Fixed navigation skill * Removed redundancy in skill training configs * Addressed some comments * Removed nav pick * Update .gitignore Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * PR comments * [v0.2.3] Hab2 fixup : merging main and resolving conflicts (facebookresearch#1022) * [v0.2.3] Add robot document (facebookresearch#1018) * [v0.2.3] [Stretch] play stretch demo (facebookresearch#1014) * [Fix] Documentation gym example path (facebookresearch#1016) Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * [0.2.3]Hydra migration 3 v2 (facebookresearch#1000) * updated md files * updated habitat2_gym_tutorial * updated Habitat2_Quickstart * updated Habitat_Lab * updated docs * partially removed YACS (note there are some inpedendent parts of code that still rely on YACS) * made all DictConfig imports to be from omegaconf using if TYPE_CHECKING * removed link to the outdated notebook * fixed tests * [Hydra Migration] Removing yacs and migratting the dataset generation code of rearrange (facebookresearch#1001) * removing yacs and migratting the dataset generation code of rearrange * addressing comments * typechecking corrections Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * renaming SensorConfig to LabSensorConfig * fixing svg link * fixing svg; * using relative path is better for svg * addressed comments * [0.2.3][Hydra Migration] Made agents to be of type Dict[str, AgentConfig] and added migration readme (facebookresearch#1007) * made habitat.simulator.agents be a Dict[str, AgentConfig] * addressed comments * refactored sim agents configuration * moved rearrange agents declarations to agents folder; finised agents configuration updates; updated docs * updated get_config import * Cfg sys migration readme (facebookresearch#2) * added habitat-lab config readme * addressed most of the comments * added main_agent alias * renamed habitat config reagme and added baselines config readme * Update habitat-lab/habitat/config/default.py Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * added desctiption for the agents_order field * added a dot :) Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * fixing conflicts Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * Fix links in MANIFEST.in (facebookresearch#991) * Fix broken links in MANIFEST.in * fixig video_render_view * fixing issue with gym maker * editing test that changed with recent config modifications * [v0.2.3] Fixing mypy (facebookresearch#1023) * Unpin habitat-lab requirements * Reorder dependencies * Fix mypy config * remove unnecessary mypy args * Update mypy version * Fix python lint * Fix pre-commit mypy further * fix mypy ignored errors * fixing mypy * fixing mypy * more accurate type:ignores * adding docs/ to exclude mypy * trying to fix python_lint * exlude arguments * exlude arguments * Container needs to be a string Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * removing the gym envs keys that were removed Co-authored-by: Jimmy Yang <55121504+jimmytyyang@users.noreply.github.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> Co-authored-by: Ruslan <partsey2412@gmail.com> Co-authored-by: Mikaël Dallaire Côté <110583667+0mdc@users.noreply.github.com> Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com> * Merged * Removed bad file and fixed IK * Env fixes * Fixed precommit * Fixed eval and fixed wandb logging * Fix for the resume in VER and the force threshold for open tasks * Update habitat-baselines/habitat_baselines/rl/ddppo/policy/resnet_policy.py Co-authored-by: Aaron Gokaslan <skylion.aaron@gmail.com> * tp-srl fixes * tp-srl fixes * Fixed more tests * Cleaned up nav action * Added tests for all tasks in baseline trainer * Corrected test overrides * Improved agents loop in resnet_policy Improved agents loop in resnet_policy Co-authored-by: Ruslan <partsey2412@gmail.com> * Fixed policy docs and improvements. Added docs to spawn robot method. * [0.2.3] 959 PR review suggested changes (facebookresearch#1029) * removed habitat-baselines/habitat_baselines/config path prefix * improved adding render view uuids to the gym obs_keys * made config keys lowercase * Updated docs * Update habitat-lab/habitat/tasks/rearrange/rearrange_task.py Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * Update habitat-baselines/habitat_baselines/rl/ddppo/policy/resnet_policy.py Co-authored-by: Aaron Gokaslan <skylion.aaron@gmail.com> * Fixed formatting * tp-srl config fix * fixing documentation ? Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> Co-authored-by: ASzot <me@andrewszot.com> Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> Co-authored-by: Jimmy Yang <55121504+jimmytyyang@users.noreply.github.com> Co-authored-by: Ruslan <partsey2412@gmail.com> Co-authored-by: Mikaël Dallaire Côté <110583667+0mdc@users.noreply.github.com> Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com> Co-authored-by: Aaron Gokaslan <skylion.aaron@gmail.com> Co-authored-by: Oleksandr <maksymets.o@gmail.com> Co-authored-by: Oleksandr Maksymets <maksymets@gmail.com> Co-authored-by: SergioArnaud <sercharnaud@gmail.com>

* updated md files * updated habitat2_gym_tutorial * updated Habitat2_Quickstart * updated Habitat_Lab * updated docs * partially removed YACS (note there are some inpedendent parts of code that still rely on YACS) * made all DictConfig imports to be from omegaconf using if TYPE_CHECKING * removed link to the outdated notebook * fixed tests * [Hydra Migration] Removing yacs and migratting the dataset generation code of rearrange (facebookresearch#1001) * removing yacs and migratting the dataset generation code of rearrange * addressing comments * typechecking corrections Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * renaming SensorConfig to LabSensorConfig * fixing svg link * fixing svg; * using relative path is better for svg * addressed comments * [0.2.3][Hydra Migration] Made agents to be of type Dict[str, AgentConfig] and added migration readme (facebookresearch#1007) * made habitat.simulator.agents be a Dict[str, AgentConfig] * addressed comments * refactored sim agents configuration * moved rearrange agents declarations to agents folder; finised agents configuration updates; updated docs * updated get_config import * Cfg sys migration readme (#2) * added habitat-lab config readme * addressed most of the comments * added main_agent alias * renamed habitat config reagme and added baselines config readme * Update habitat-lab/habitat/config/default.py Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * added desctiption for the agents_order field * added a dot :) Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * fixing conflicts Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com>

…okresearch#959) * [bugfix] Reconfigure the sim even if the sim is not None * addressing comments, better way to decide wether or not to reconfigure sim_obj * same fix for the robot base * Added back caching to pick task * Fixed up nav to obj task * Fixed up open and close skills * Added back inits for nav to obj * Batched HRL policy * Updated task conditions * Fixed key mismatch in TP-SRL * Fixed issue with nav to obj skill * Fixed early stop issue * Fixed bad filter condition * Updated training configs * Refactored and cleaned up config files * Fixed up reward. Renamed configs * Added wandb resume * Refactored eval with debug sensor code. * Added get policy info * Hopefully fixed resuming, added safe drop * Added improved Fetch model * Removed refactored files * Added back default * Removed habitat folder * Fixed pick task * Fixed navigation skill * Removed redundancy in skill training configs * Addressed some comments * Removed nav pick * Update .gitignore Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * PR comments * [v0.2.3] Hab2 fixup : merging main and resolving conflicts (facebookresearch#1022) * [v0.2.3] Add robot document (facebookresearch#1018) * [v0.2.3] [Stretch] play stretch demo (facebookresearch#1014) * [Fix] Documentation gym example path (facebookresearch#1016) Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * [0.2.3]Hydra migration 3 v2 (facebookresearch#1000) * updated md files * updated habitat2_gym_tutorial * updated Habitat2_Quickstart * updated Habitat_Lab * updated docs * partially removed YACS (note there are some inpedendent parts of code that still rely on YACS) * made all DictConfig imports to be from omegaconf using if TYPE_CHECKING * removed link to the outdated notebook * fixed tests * [Hydra Migration] Removing yacs and migratting the dataset generation code of rearrange (facebookresearch#1001) * removing yacs and migratting the dataset generation code of rearrange * addressing comments * typechecking corrections Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * renaming SensorConfig to LabSensorConfig * fixing svg link * fixing svg; * using relative path is better for svg * addressed comments * [0.2.3][Hydra Migration] Made agents to be of type Dict[str, AgentConfig] and added migration readme (facebookresearch#1007) * made habitat.simulator.agents be a Dict[str, AgentConfig] * addressed comments * refactored sim agents configuration * moved rearrange agents declarations to agents folder; finised agents configuration updates; updated docs * updated get_config import * Cfg sys migration readme (#2) * added habitat-lab config readme * addressed most of the comments * added main_agent alias * renamed habitat config reagme and added baselines config readme * Update habitat-lab/habitat/config/default.py Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * added desctiption for the agents_order field * added a dot :) Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * fixing conflicts Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * Fix links in MANIFEST.in (facebookresearch#991) * Fix broken links in MANIFEST.in * fixig video_render_view * fixing issue with gym maker * editing test that changed with recent config modifications * [v0.2.3] Fixing mypy (facebookresearch#1023) * Unpin habitat-lab requirements * Reorder dependencies * Fix mypy config * remove unnecessary mypy args * Update mypy version * Fix python lint * Fix pre-commit mypy further * fix mypy ignored errors * fixing mypy * fixing mypy * more accurate type:ignores * adding docs/ to exclude mypy * trying to fix python_lint * exlude arguments * exlude arguments * Container needs to be a string Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> * removing the gym envs keys that were removed Co-authored-by: Jimmy Yang <55121504+jimmytyyang@users.noreply.github.com> Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> Co-authored-by: Ruslan <partsey2412@gmail.com> Co-authored-by: Mikaël Dallaire Côté <110583667+0mdc@users.noreply.github.com> Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com> * Merged * Removed bad file and fixed IK * Env fixes * Fixed precommit * Fixed eval and fixed wandb logging * Fix for the resume in VER and the force threshold for open tasks * Update habitat-baselines/habitat_baselines/rl/ddppo/policy/resnet_policy.py Co-authored-by: Aaron Gokaslan <skylion.aaron@gmail.com> * tp-srl fixes * tp-srl fixes * Fixed more tests * Cleaned up nav action * Added tests for all tasks in baseline trainer * Corrected test overrides * Improved agents loop in resnet_policy Improved agents loop in resnet_policy Co-authored-by: Ruslan <partsey2412@gmail.com> * Fixed policy docs and improvements. Added docs to spawn robot method. * [0.2.3] 959 PR review suggested changes (facebookresearch#1029) * removed habitat-baselines/habitat_baselines/config path prefix * improved adding render view uuids to the gym obs_keys * made config keys lowercase * Updated docs * Update habitat-lab/habitat/tasks/rearrange/rearrange_task.py Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> * Update habitat-baselines/habitat_baselines/rl/ddppo/policy/resnet_policy.py Co-authored-by: Aaron Gokaslan <skylion.aaron@gmail.com> * Fixed formatting * tp-srl config fix * fixing documentation ? Co-authored-by: vincentpierre <vincentpierre@users.noreply.github.com> Co-authored-by: ASzot <me@andrewszot.com> Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com> Co-authored-by: Jimmy Yang <55121504+jimmytyyang@users.noreply.github.com> Co-authored-by: Ruslan <partsey2412@gmail.com> Co-authored-by: Mikaël Dallaire Côté <110583667+0mdc@users.noreply.github.com> Co-authored-by: Aaron Gokaslan <aaronGokaslan@gmail.com> Co-authored-by: Aaron Gokaslan <skylion.aaron@gmail.com> Co-authored-by: Oleksandr <maksymets.o@gmail.com> Co-authored-by: Oleksandr Maksymets <maksymets@gmail.com> Co-authored-by: SergioArnaud <sercharnaud@gmail.com>

made habitat.simulator.agents be a Dict[str, AgentConfig]

af48d72

rpartsey requested a review from vincentpierre November 19, 2022 17:39

facebook-github-bot added the CLA Signed Do not delete this pull request or issue due to inactivity. label Nov 19, 2022

vincentpierre approved these changes Nov 21, 2022

View reviewed changes

vincentpierre reviewed Nov 21, 2022

View reviewed changes

Skylion007 approved these changes Nov 22, 2022

View reviewed changes

addressed comments

c82cdb4

rpartsey requested a review from vincentpierre November 23, 2022 21:49

rpartsey added 3 commits November 26, 2022 11:48

refactored sim agents configuration

b56447e

moved rearrange agents declarations to agents folder; finised agents …

bf44e6d

…configuration updates; updated docs

updated get_config import

5eb9ef2

rpartsey commented Nov 27, 2022

View reviewed changes

Cfg sys migration readme (#2)

cf0100e

* added habitat-lab config readme

vincentpierre changed the title ~~[Hydra Migration] Made habitat.simulator.agents to be of type Dict[str, AgentConfig]~~ [0.2.3][Hydra Migration] Made habitat.simulator.agents to be of type Dict[str, AgentConfig] Nov 28, 2022

rpartsey changed the title ~~[0.2.3][Hydra Migration] Made habitat.simulator.agents to be of type Dict[str, AgentConfig]~~ [0.2.3][Hydra Migration] Made agents to be of type Dict[str, AgentConfig] and added migration readme Nov 28, 2022

vincentpierre mentioned this pull request Nov 28, 2022

[0.2.3]Hydra migration 3 v2 #1000

Merged

7 tasks

vincentpierre reviewed Nov 28, 2022

View reviewed changes

vincentpierre reviewed Nov 29, 2022

View reviewed changes

vincentpierre approved these changes Nov 29, 2022

View reviewed changes

mathfac approved these changes Nov 29, 2022

View reviewed changes

mathfac reviewed Nov 29, 2022

View reviewed changes

mathfac approved these changes Nov 29, 2022

View reviewed changes

mathfac reviewed Nov 29, 2022

View reviewed changes

rpartsey added 3 commits November 29, 2022 17:14

addressed most of the comments

d05b5b9

added main_agent alias

b8d32dd

renamed habitat config reagme and added baselines config readme

87fdd95

rpartsey requested review from mathfac and vincentpierre November 29, 2022 17:06

vincentpierre approved these changes Nov 29, 2022

View reviewed changes

Merge branch 'hydra_migration_3_v2' into hydra_migration_3_v2_agents_upd

fcffeb3

Skylion007 reviewed Nov 29, 2022

View reviewed changes

Skylion007 suggested changes Nov 29, 2022

View reviewed changes

Skylion007 reviewed Nov 29, 2022

View reviewed changes

Skylion007 approved these changes Nov 29, 2022

View reviewed changes

rpartsey and others added 4 commits November 30, 2022 01:00

Update habitat-lab/habitat/config/default.py

de5150e

Co-authored-by: Vincent-Pierre BERGES <28320361+vincentpierre@users.noreply.github.com>

added desctiption for the agents_order field

da80fd3

Merge branch 'hydra_migration_3_v2_agents_upd' of github.com:rpartsey…

66cbd8c

…/habitat-lab into hydra_migration_3_v2_agents_upd

added a dot :)

ce2ec58

vincentpierre merged commit b40736f into facebookresearch:hydra_migration_3_v2 Nov 29, 2022

	custom_config_key: KeyType = DefayltValue
	custom_config_key: KeyType = DefaultValue

	is composed dynamically at the run time from
	is composed dynamically at run time from

	Structured Configs registry. Habitat-Lab's Structured Configs are defined and registered to the ConfigStore in the
	Structured Configs registry. Habitat-Lab's Structured Configs are defined and registered to the ConfigStore in

	Similar configs are grouped in the [Config Groups](https://hydra.cc/docs/advanced/terminology/#config-group)
	Similar configs are grouped in [Config Groups](https://hydra.cc/docs/advanced/terminology/#config-group)

	Where the content of each Input Config (Config Group Option) is placed in the Output Config is determined via
	Each Input Config's position in the Output Config is determined via

	<summary>Expand to see the Habitat-Lab input PointNav benchmark config example.</summary>
	PointNav benchmark: Input config.</summary>

	<summary>Expand to see the corresponding composed output config.</summary>
	<summary>
	PointNav benchmark: Output config.</summary>

	`simulator.agent_name.sim_sensors` sub-nodes of type `Dict[str, CorrespondingConfigNodeType]`
	`simulator.<agent_name>.sim_sensors` sub-nodes of type `Dict[str, CorrespondingConfigNodeType]`

	"habitat.simulator.agents_order should be set explicitly "
	"habitat.simulator.agents_order({sim_config.agents_order}) should be set explicitly "

	"and match the agents' keys in habitat.simulator.agents"
	"and match the agents' keys in habitat.simulator.agents({set(sim_config.agents.keys())})"

		# if the number of agents is greater than one,
		# then agents_order has to be set explicitly too

	# if the number of agents is greater than one,
	# then agents_order has to be set explicitly too
	# Agents order that will be used by simulator step and trainer policy.

		)


		def get_agent_config(sim_config, agent_id: Optional[int] = None):

-def get_agent_config(sim_config, agent_id: Optional[int] = None):
+def get_agent_config(sim_config: DictConfig, agent_id: Optional[int] = None) -> DictConfig:
+        r"""Called to get agent's config node of default agent or based on index of the agent.
+        :param sim_config: config of :ref:`core.simulator.Simulator`.
+        :param agent_id: index of the agent (relevant for multi-agent setup).
+        :return: relevant agent's config.
+        """

	agent_config = get_agent_config(config.habitat.simulator)
	agent_config = get_agent_config(sim_config=config.habitat.simulator)

	- [Multirun](https://hydra.cc/docs/tutorials/basic/running_your_app/multi-run/). For example, launching 3 experiments with three different learning rates:
	- [Parameter sweeping and multirun](https://hydra.cc/docs/tutorials/basic/running_your_app/multi-run/). For example, launching 3 experiments with three different learning rates:

+- Seamless [SLURM](https://slurm.schedmd.com/documentation.html) integration through [Submitit Launcher](https://hydra.cc/docs/plugins/submitit_launcher/). To enable the feature Submitit plugin should be installed: `pip install hydra-submitit-launcher --upgrade` and specify launcher `hydra/launcher=submitit_slurm` in the command line:
+```bash
+python -u habitat_baselines/run.py --exp-config config.yaml --run-type train –-multirun \
+habitat_baselines.rl.ppo.lr 2.5e-4,2.5e-5,2.5e-6 hydra/launcher=submitit_slurm

		@@ -0,0 +1,5 @@
		# @package habitat.simulator.agents.depth_agent

	agents: Dict[str, AgentConfig] = MISSING
	agents: List[AgentConfig] = MISSING

	agents: Dict[str, AgentConfig] = MISSING
	agents: OrderedDict[str, AgentConfig] = MISSING

[0.2.3][Hydra Migration] Made agents to be of type Dict[str, AgentConfig] and added migration readme #1007

[0.2.3][Hydra Migration] Made agents to be of type Dict[str, AgentConfig] and added migration readme #1007

Conversation

rpartsey commented Nov 19, 2022 • edited Loading

Motivation and Context

How Has This Been Tested

Types of changes

Checklist

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mathfac Nov 29, 2022 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Skylion007 Nov 29, 2022 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Skylion007 Nov 29, 2022 • edited Loading

Choose a reason for hiding this comment

Skylion007 left a comment • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Skylion007 Nov 29, 2022 • edited Loading

Choose a reason for hiding this comment

Skylion007 left a comment

Choose a reason for hiding this comment

rpartsey commented Nov 19, 2022 •

edited

Loading

mathfac Nov 29, 2022 •

edited

Loading

Skylion007 Nov 29, 2022 •

edited

Loading

Skylion007 Nov 29, 2022 •

edited

Loading

Skylion007 left a comment •

edited

Loading

Skylion007 Nov 29, 2022 •

edited

Loading