tangger/lerobot_piper
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

fix environment seeding

Browse Source 
add fixes for reproducibility

only try to start env if it is closed

revision

fix normalization and data type

Improve README

Improve README

Tests are passing, Eval pretrained model works, Add gif

Update gif

Update gif

Update gif

Update gif

Update README

Update README

update minor

Update README.md

Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>

Update README.md

Co-authored-by: Simon Alibert <75076266+aliberts@users.noreply.github.com>

Address suggestions

Update thumbnail + stats

Update thumbnail + stats

Update README.md

Co-authored-by: Alexander Soare <alexander.soare159@gmail.com>

Add more comments

Add test_examples.py
This commit is contained in:
Alexander Soare
2024-03-22 13:25:23 +00:00
committed by Cadene
parent 203bcd7ca5
commit 1a1308d62f
32 changed files with 686 additions and 282 deletions
Download Patch File Download Diff File Expand all files Collapse all files

46
lerobot/common/datasets/abstract.py
View File

@@ -9,7 +9,7 @@ import tqdm
from huggingface_hub import snapshot_download
from tensordict import TensorDict
from torchrl.data.replay_buffers.replay_buffers import TensorDictReplayBuffer
from torchrl.data.replay_buffers.samplers import SliceSampler
from torchrl.data.replay_buffers.samplers import Sampler
from torchrl.data.replay_buffers.storages import TensorStorage, _collate_id
from torchrl.data.replay_buffers.writers import ImmutableDatasetWriter, Writer
from torchrl.envs.transforms.transforms import Compose
@@ -17,22 +17,56 @@ from torchrl.envs.transforms.transforms import Compose
HF_USER = "lerobot"
class AbstractExperienceReplay(TensorDictReplayBuffer):
class AbstractDataset(TensorDictReplayBuffer):
"""
AbstractDataset represents a dataset in the context of imitation learning or reinforcement learning.
This class is designed to be subclassed by concrete implementations that specify particular types of datasets.
These implementations can vary based on the source of the data, the environment the data pertains to,
or the specific kind of data manipulation applied.
Note:
- `TensorDictReplayBuffer` is the base class from which `AbstractDataset` inherits. It provides the foundational
functionality for storing and retrieving `TensorDict`-like data.
- `available_datasets` should be overridden by concrete subclasses to list the specific dataset variants supported.
It is expected that these variants correspond to a HuggingFace dataset on the hub.
For instance, the `AlohaDataset` which inherites from `AbstractDataset` has 4 available dataset variants:
- [aloha_sim_transfer_cube_scripted](https://huggingface.co/datasets/lerobot/aloha_sim_transfer_cube_scripted)
- [aloha_sim_insertion_scripted](https://huggingface.co/datasets/lerobot/aloha_sim_insertion_scripted)
- [aloha_sim_transfer_cube_human](https://huggingface.co/datasets/lerobot/aloha_sim_transfer_cube_human)
- [aloha_sim_insertion_human](https://huggingface.co/datasets/lerobot/aloha_sim_insertion_human)
- When implementing a concrete class (e.g. `AlohaDataset`, `PushtEnv`, `DiffusionPolicy`), you need to:
1. set the required class attributes:
- for classes inheriting from `AbstractDataset`: `available_datasets`
- for classes inheriting from `AbstractEnv`: `name`, `available_tasks`
- for classes inheriting from `AbstractPolicy`: `name`
2. update variables in `lerobot/__init__.py` (e.g. `available_envs`, `available_datasets_per_envs`, `available_policies`)
3. update variables in `tests/test_available.py` by importing your new class
"""
available_datasets: list[str] | None = None
def __init__(
self,
dataset_id: str,
version: str | None = None,
batch_size: int = None,
batch_size: int | None = None,
*,
shuffle: bool = True,
root: Path | None = None,
pin_memory: bool = False,
prefetch: int = None,
sampler: SliceSampler = None,
collate_fn: Callable = None,
writer: Writer = None,
sampler: Sampler | None = None,
collate_fn: Callable | None = None,
writer: Writer | None = None,
transform: "torchrl.envs.Transform" = None,
):
assert (
self.available_datasets is not None
), "Subclasses of `AbstractDataset` should set the `available_datasets` class attribute."
assert (
dataset_id in self.available_datasets
), f"The provided dataset ({dataset_id}) is not on the list of available datasets {self.available_datasets}."
self.dataset_id = dataset_id
self.version = version
self.shuffle = shuffle

18
lerobot/common/datasets/aloha.py
View File

@@ -9,11 +9,11 @@ import torch
import torchrl
import tqdm
from tensordict import TensorDict
from torchrl.data.replay_buffers.samplers import SliceSampler
from torchrl.data.replay_buffers.samplers import Sampler
from torchrl.data.replay_buffers.storages import TensorStorage
from torchrl.data.replay_buffers.writers import Writer
from lerobot.common.datasets.abstract import AbstractExperienceReplay
from lerobot.common.datasets.abstract import AbstractDataset
DATASET_IDS = [
"aloha_sim_insertion_human",
@@ -80,24 +80,24 @@ def download(data_dir, dataset_id):
gdown.download(EP49_URLS[dataset_id], output=str(data_dir / "episode_49.hdf5"), fuzzy=True)
class AlohaExperienceReplay(AbstractExperienceReplay):
class AlohaDataset(AbstractDataset):
available_datasets = DATASET_IDS
def __init__(
self,
dataset_id: str,
version: str | None = "v1.2",
batch_size: int = None,
batch_size: int | None = None,
*,
shuffle: bool = True,
root: Path | None = None,
pin_memory: bool = False,
prefetch: int = None,
sampler: SliceSampler = None,
collate_fn: Callable = None,
writer: Writer = None,
sampler: Sampler | None = None,
collate_fn: Callable | None = None,
writer: Writer | None = None,
transform: "torchrl.envs.Transform" = None,
):
assert dataset_id in DATASET_IDS
super().__init__(
dataset_id,
version,

75
lerobot/common/datasets/factory.py
View File

@@ -5,7 +5,7 @@ from pathlib import Path
import torch
from torchrl.data.replay_buffers import PrioritizedSliceSampler, SliceSampler
from lerobot.common.envs.transforms import NormalizeTransform, Prod
from lerobot.common.transforms import NormalizeTransform, Prod
# DATA_DIR specifies to location where datasets are loaded. By default, DATA_DIR is None and
# we load from `$HOME/.cache/huggingface/hub/datasets`. For our unit tests, we set `DATA_DIR=tests/data`
@@ -16,6 +16,7 @@ DATA_DIR = Path(os.environ["DATA_DIR"]) if "DATA_DIR" in os.environ else None
def make_offline_buffer(
cfg,
overwrite_sampler=None,
# set normalize=False to remove all transformations and keep images unnormalized in [0,255]
normalize=True,
overwrite_batch_size=None,
overwrite_prefetch=None,
@@ -64,25 +65,27 @@ def make_offline_buffer(
sampler = overwrite_sampler
if cfg.env.name == "simxarm":
from lerobot.common.datasets.simxarm import SimxarmExperienceReplay
from lerobot.common.datasets.simxarm import SimxarmDataset
clsfunc = SimxarmExperienceReplay
dataset_id = f"xarm_{cfg.env.task}_medium"
clsfunc = SimxarmDataset
elif cfg.env.name == "pusht":
from lerobot.common.datasets.pusht import PushtExperienceReplay
from lerobot.common.datasets.pusht import PushtDataset
clsfunc = PushtExperienceReplay
dataset_id = "pusht"
clsfunc = PushtDataset
elif cfg.env.name == "aloha":
from lerobot.common.datasets.aloha import AlohaExperienceReplay
from lerobot.common.datasets.aloha import AlohaDataset
clsfunc = AlohaExperienceReplay
dataset_id = f"aloha_{cfg.env.task}"
clsfunc = AlohaDataset
else:
raise ValueError(cfg.env.name)
# TODO(rcadene): backward compatiblity to load pretrained pusht policy
dataset_id = cfg.get("dataset_id")
if dataset_id is None and cfg.env.name == "pusht":
dataset_id = "pusht"
offline_buffer = clsfunc(
dataset_id=dataset_id,
sampler=sampler,
@@ -100,36 +103,40 @@ def make_offline_buffer(
else:
img_keys = offline_buffer.image_keys
transforms = [Prod(in_keys=img_keys, prod=1 / 255)]
if normalize:
transforms = [Prod(in_keys=img_keys, prod=1 / 255)]
if normalize:
# TODO(rcadene): make normalization strategy configurable between mean_std, min_max, manual_min_max,
# min_max_from_spec
stats = offline_buffer.compute_or_load_stats() if stats_path is None else torch.load(stats_path)
# TODO(rcadene): make normalization strategy configurable between mean_std, min_max, manual_min_max,
# min_max_from_spec
stats = offline_buffer.compute_or_load_stats() if stats_path is None else torch.load(stats_path)
# we only normalize the state and action, since the images are usually normalized inside the model for
# now (except for tdmpc: see the following)
in_keys = [("observation", "state"), ("action")]
# we only normalize the state and action, since the images are usually normalized inside the model for
# now (except for tdmpc: see the following)
in_keys = [("observation", "state"), ("action")]
if cfg.policy.name == "tdmpc":
# TODO(rcadene): we add img_keys to the keys to normalize for tdmpc only, since diffusion and act policies normalize the image inside the model for now
in_keys += img_keys
# TODO(racdene): since we use next observations in tdmpc, we also add them to the normalization. We are wasting a bit of compute on this for now.
in_keys += [("next", *key) for key in img_keys]
in_keys.append(("next", "observation", "state"))
if cfg.policy.name == "tdmpc":
# TODO(rcadene): we add img_keys to the keys to normalize for tdmpc only, since diffusion and act policies normalize the image inside the model for now
in_keys += img_keys
# TODO(racdene): since we use next observations in tdmpc, we also add them to the normalization. We are wasting a bit of compute on this for now.
in_keys += [("next", *key) for key in img_keys]
in_keys.append(("next", "observation", "state"))
if cfg.policy.name == "diffusion" and cfg.env.name == "pusht":
# TODO(rcadene): we overwrite stats to have the same as pretrained model, but we should remove this
stats["observation", "state", "min"] = torch.tensor([13.456424, 32.938293], dtype=torch.float32)
stats["observation", "state", "max"] = torch.tensor([496.14618, 510.9579], dtype=torch.float32)
stats["action", "min"] = torch.tensor([12.0, 25.0], dtype=torch.float32)
stats["action", "max"] = torch.tensor([511.0, 511.0], dtype=torch.float32)
if cfg.policy.name == "diffusion" and cfg.env.name == "pusht":
# TODO(rcadene): we overwrite stats to have the same as pretrained model, but we should remove this
stats["observation", "state", "min"] = torch.tensor(
[13.456424, 32.938293], dtype=torch.float32
)
stats["observation", "state", "max"] = torch.tensor(
[496.14618, 510.9579], dtype=torch.float32
)
stats["action", "min"] = torch.tensor([12.0, 25.0], dtype=torch.float32)
stats["action", "max"] = torch.tensor([511.0, 511.0], dtype=torch.float32)
# TODO(rcadene): remove this and put it in config. Ideally we want to reproduce SOTA results just with mean_std
normalization_mode = "mean_std" if cfg.env.name == "aloha" else "min_max"
transforms.append(NormalizeTransform(stats, in_keys, mode=normalization_mode))
# TODO(rcadene): remove this and put it in config. Ideally we want to reproduce SOTA results just with mean_std
normalization_mode = "mean_std" if cfg.env.name == "aloha" else "min_max"
transforms.append(NormalizeTransform(stats, in_keys, mode=normalization_mode))
offline_buffer.set_transform(transforms)
offline_buffer.set_transform(transforms)
if not overwrite_sampler:
index = torch.arange(0, offline_buffer.num_samples, 1)

16
lerobot/common/datasets/pusht.py
View File

@@ -9,11 +9,11 @@ import torch
import torchrl
import tqdm
from tensordict import TensorDict
from torchrl.data.replay_buffers.samplers import SliceSampler
from torchrl.data.replay_buffers.samplers import Sampler
from torchrl.data.replay_buffers.storages import TensorStorage
from torchrl.data.replay_buffers.writers import Writer
from lerobot.common.datasets.abstract import AbstractExperienceReplay
from lerobot.common.datasets.abstract import AbstractDataset
from lerobot.common.datasets.utils import download_and_extract_zip
from lerobot.common.envs.pusht.pusht_env import pymunk_to_shapely
from lerobot.common.policies.diffusion.replay_buffer import ReplayBuffer as DiffusionPolicyReplayBuffer
@@ -83,20 +83,22 @@ def add_tee(
return body
class PushtExperienceReplay(AbstractExperienceReplay):
class PushtDataset(AbstractDataset):
available_datasets = ["pusht"]
def __init__(
self,
dataset_id: str,
version: str | None = "v1.2",
batch_size: int = None,
batch_size: int | None = None,
*,
shuffle: bool = True,
root: Path | None = None,
pin_memory: bool = False,
prefetch: int = None,
sampler: SliceSampler = None,
collate_fn: Callable = None,
writer: Writer = None,
sampler: Sampler | None = None,
collate_fn: Callable | None = None,
writer: Writer | None = None,
transform: "torchrl.envs.Transform" = None,
):
super().__init__(

14
lerobot/common/datasets/simxarm.py
View File

@@ -8,12 +8,12 @@ import torchrl
import tqdm
from tensordict import TensorDict
from torchrl.data.replay_buffers.samplers import (
SliceSampler,
Sampler,
)
from torchrl.data.replay_buffers.storages import TensorStorage
from torchrl.data.replay_buffers.writers import Writer
from lerobot.common.datasets.abstract import AbstractExperienceReplay
from lerobot.common.datasets.abstract import AbstractDataset
def download():
@@ -32,7 +32,7 @@ def download():
Path(download_path).unlink()
class SimxarmExperienceReplay(AbstractExperienceReplay):
class SimxarmDataset(AbstractDataset):
available_datasets = [
"xarm_lift_medium",
]
@@ -41,15 +41,15 @@ class SimxarmExperienceReplay(AbstractExperienceReplay):
self,
dataset_id: str,
version: str | None = "v1.1",
batch_size: int = None,
batch_size: int | None = None,
*,
shuffle: bool = True,
root: Path | None = None,
pin_memory: bool = False,
prefetch: int = None,
sampler: SliceSampler = None,
collate_fn: Callable = None,
writer: Writer = None,
sampler: Sampler | None = None,
collate_fn: Callable | None = None,
writer: Writer | None = None,
transform: "torchrl.envs.Transform" = None,
):
super().__init__(

22
lerobot/common/envs/abstract.py
View File

@@ -8,6 +8,20 @@ from lerobot.common.utils import set_global_seed
class AbstractEnv(EnvBase):
"""
Note:
When implementing a concrete class (e.g. `AlohaDataset`, `PushtEnv`, `DiffusionPolicy`), you need to:
1. set the required class attributes:
- for classes inheriting from `AbstractDataset`: `available_datasets`
- for classes inheriting from `AbstractEnv`: `name`, `available_tasks`
- for classes inheriting from `AbstractPolicy`: `name`
2. update variables in `lerobot/__init__.py` (e.g. `available_envs`, `available_datasets_per_envs`, `available_policies`)
3. update variables in `tests/test_available.py` by importing your new class
"""
name: str | None = None # same name should be used to instantiate the environment in factory.py
available_tasks: list[str] | None = None # for instance: sim_insertion, sim_transfer_cube, pusht, lift
def __init__(
self,
task,
@@ -21,6 +35,14 @@ class AbstractEnv(EnvBase):
num_prev_action=0,
):
super().__init__(device=device, batch_size=[])
assert self.name is not None, "Subclasses of `AbstractEnv` should set the `name` class attribute."
assert (
self.available_tasks is not None
), "Subclasses of `AbstractEnv` should set the `available_tasks` class attribute."
assert (
task in self.available_tasks
), f"The provided task ({task}) is not on the list of available tasks {self.available_tasks}."
self.task = task
self.frame_skip = frame_skip
self.from_pixels = from_pixels

2
lerobot/common/envs/aloha/env.py
View File

@@ -35,6 +35,8 @@ _has_gym = importlib.util.find_spec("gymnasium") is not None
class AlohaEnv(AbstractEnv):
name = "aloha"
available_tasks = ["sim_insertion", "sim_transfer_cube"]
_reset_warning_issued = False
def __init__(

2
lerobot/common/envs/pusht/env.py
View File

@@ -22,6 +22,8 @@ _has_gym = importlib.util.find_spec("gymnasium") is not None
class PushtEnv(AbstractEnv):
name = "pusht"
available_tasks = ["pusht"]
_reset_warning_issued = False
def __init__(

3
lerobot/common/envs/simxarm/env.py
View File

@@ -24,6 +24,9 @@ _has_gym = importlib.util.find_spec("gymnasium") is not None
class SimxarmEnv(AbstractEnv):
name = "simxarm"
available_tasks = ["lift"]
def __init__(
self,
task,

12
lerobot/common/policies/abstract.py
View File

@@ -9,8 +9,19 @@ class AbstractPolicy(nn.Module):
The forward method should generally not be overriden as it plays the role of handling multi-step policies. See its
documentation for more information.
Note:
When implementing a concrete class (e.g. `AlohaDataset`, `PushtEnv`, `DiffusionPolicy`), you need to:
1. set the required class attributes:
- for classes inheriting from `AbstractDataset`: `available_datasets`
- for classes inheriting from `AbstractEnv`: `name`, `available_tasks`
- for classes inheriting from `AbstractPolicy`: `name`
2. update variables in `lerobot/__init__.py` (e.g. `available_envs`, `available_datasets_per_envs`, `available_policies`)
3. update variables in `tests/test_available.py` by importing your new class
"""
name: str | None = None # same name should be used to instantiate the policy in factory.py
def __init__(self, n_action_steps: int | None):
"""
n_action_steps: Sets the cache size for storing action trajectories. If None, it is assumed that a single
@@ -18,6 +29,7 @@ class AbstractPolicy(nn.Module):
adds that dimension.
"""
super().__init__()
assert self.name is not None, "Subclasses of `AbstractPolicy` should set the `name` class attribute."
self.n_action_steps = n_action_steps
self.clear_action_queue()

2
lerobot/common/policies/act/policy.py
View File

@@ -42,6 +42,8 @@ def kl_divergence(mu, logvar):
class ActionChunkingTransformerPolicy(AbstractPolicy):
name = "act"
def __init__(self, cfg, device, n_action_steps=1):
super().__init__(n_action_steps)
self.cfg = cfg

2
lerobot/common/policies/diffusion/policy.py
View File

@@ -13,6 +13,8 @@ from lerobot.common.utils import get_safe_torch_device
class DiffusionPolicy(AbstractPolicy):
name = "diffusion"
def __init__(
self,
cfg,

4
lerobot/common/policies/factory.py
View File

@@ -3,9 +3,9 @@ def make_policy(cfg):
raise NotImplementedError("Only diffusion policy supports rollout_batch_size > 1 for the time being.")
if cfg.policy.name == "tdmpc":
from lerobot.common.policies.tdmpc.policy import TDMPC
from lerobot.common.policies.tdmpc.policy import TDMPCPolicy
policy = TDMPC(cfg.policy, cfg.device)
policy = TDMPCPolicy(cfg.policy, cfg.device)
elif cfg.policy.name == "diffusion":
from lerobot.common.policies.diffusion.policy import DiffusionPolicy

4
lerobot/common/policies/tdmpc/policy.py
View File

@@ -87,9 +87,11 @@ class TOLD(nn.Module):
return torch.min(Q1, Q2) if return_type == "min" else (Q1 + Q2) / 2
class TDMPC(AbstractPolicy):
class TDMPCPolicy(AbstractPolicy):
"""Implementation of TD-MPC learning + inference."""
name = "tdmpc"
def __init__(self, cfg, device):
super().__init__(None)
self.action_dim = cfg.action_dim

0
lerobot/common/envs/transforms.py → lerobot/common/transforms.py
View File