Add OpenPi, Pi0 and Pi0.5 (#1910)
* initial commit * change device in test * do detailed import * adhere to python 3.11 syntax * fix autodocstring * additionally * do same in other files * add model. prefix to all keys in state dict * use dummy stats * add pi05 * also shorten action_steps * fix test * all test pass! and fix tokenizer max length between 05 and 0 * remove test * fix transformer dependency * fix test * split pi0 and pi05 policy in seperate files * fix test * fix push to hub test * add some comments, license and readme * remove warning in config * add pi05 to factory * remove check * rename action_horizon to chunk_size * clean up padding of state and action (more in line with lerobot pi0) * add openpi image transforms for training and add more flexibility to _preprocess_images similar to lerobot pi0 * fix key match from pytorch state dict (similar keys to openpi implementation now) * also for pi05 * update to python 3.11 * revert to openpi transformer replace python 3.11 * fix(modeling pi0): nit warning message * use safeauto_docstring * fix: remove unused param * fix from pretrained * add preprocess tests * also compile forward method * Do not add model prefix to normalization * use same name for action and state dim as lerobot pi0 and remove fixed image keys * load from pretrained_path * temp: hardcode base model * fix override self.pretrained_path = None overwrite * rename to loss * remove additional image augmentations, lerobot dataset already does this * Add docs * put tests in test folder * Add test to instatiate all base models * go back to python 3.10 * update docs * adapt docs pi05 * change docs: finetune base model options * minor docs fixes and dependencies * remove todo * cast float64 to float32 for mps * skip if no transformers * fix tests * add new models to modelcard * add back init * fix circular input * feat: only run pi test on GPU * remove require_nightly_gpu * replace decorator test_pi0_openpi * rename action_dim, state_dim to max_action_dim, max_state_dim * fix doc and constants * cleanup tests * fix from pretrained * fix tests * add comment pi0 pi05 tests, add image features to pi0 pi05 hub tests * fix, state is included in language not in flow head * Move test to specific folder * and paligemma task with newline * remove add_special_tokens, not needed * feedback pr * Remove previous pi0 and rename pi0_openpi and pi05_openpi * Add Quantile stats to LeRobotDataset (#1985) * - Add RunningQuantileStats class for efficient histogram-based quantile computation - Integrate quantile parameters (compute_quantiles, quantiles) into LeRobotDataset - Support quantile computation during episode collection and aggregation - Add comprehensive function-based test suite (24 tests) for quantile functionality - Maintain full backward compatibility with existing stats computation - Enable configurable quantiles (default: [0.01, 0.99]) for robust normalization * style fixes, make quantiles computation by default to new datasets * fix tests * - Added DEFAULT_QUANTILES=[0.01, 0.10, 0.50, 0.90, 0.99] to be computed for each features instead of being chosen by the user - Fortified tests. * - add helper functions to reshape stats - add missing test for quantiles * - Add QUANTILE normalization mode to normalize the data with the 1st and 99th percentiles. - Add QUANTILE10 normalization mode to normalize the data with the 10th and 90th percentiles. * style fixes * Added missing lisence * Simplify compute_stats * - added script `augment_dataset_quantile_stats.py` so that we can add quantile stats to existing v3 datasets that dont have quatniles - modified quantile computation instead of using the edge for the value, interpolate the values in the bin * rename pi0/pi05 files * Remove open pi patch and use custom transformer branch for now * renaming * fix * Revert "fix" This reverts commit 1ea65730ac2cbca6e5869df734fbd4392561b3c6. * fix naming * feet(pi0/pi0.5): add pipeline (#2009) * feat(processor): convert openpi model with processor * TODO: Make test works * fix(modeling_pi0openpi): update attention mask value and time scaling; improve task handling in tests - Changed the attention mask value from `self.config.attention_mask_value` to a fixed value of `-2.3819763e38`. - Updated time scaling in the `sample_noise` method to use a constant factor of `0.999` and an offset of `0.001`. - Enhanced task handling in tests to ensure proper formatting and batch size consistency. - Cleaned up commented-out test code for clarity. * refactor(pi0): rename PI0OpenPIConfig and PI0OpenPIPolicy to PI0Config and PI0Policy - Updated imports and references throughout the codebase to reflect the new naming convention. - Introduced a new processor file for PI0 to handle pre-processing and post-processing steps. - Adjusted tests to utilize the renamed classes, ensuring consistency and functionality. - Enhanced clarity and maintainability by removing outdated naming conventions. * refactor(pi05): rename PI0OpenPIPolicy to PI0Policy and update configuration - Renamed `PI0OpenPIPolicy` to `PI0Policy` for consistency with naming conventions. - Updated the `PI05OpenPIConfig` to include a new `tokenizer_max_length` attribute and changed the normalization mode for state from `MEAN_STD` to `QUANTILES`. - Simplified model initialization in `PI05OpenPIPolicy` by removing unused `dataset_stats` parameter. - Added a new processor class for `Pi05PrepareStateTokenizerProcessorStep` with `@dataclass` for improved readability. - Introduced a test script to compare the integration of the PI0OpenPI policy with the original implementation, ensuring local testing compatibility. * feat(processor): convert openpi model with processor * TODO: Make test works * fix(modeling_pi0openpi): update attention mask value and time scaling; improve task handling in tests - Changed the attention mask value from `self.config.attention_mask_value` to a fixed value of `-2.3819763e38`. - Updated time scaling in the `sample_noise` method to use a constant factor of `0.999` and an offset of `0.001`. - Enhanced task handling in tests to ensure proper formatting and batch size consistency. - Cleaned up commented-out test code for clarity. * refactor(pi0): rename PI0OpenPIConfig and PI0OpenPIPolicy to PI0Config and PI0Policy - Updated imports and references throughout the codebase to reflect the new naming convention. - Introduced a new processor file for PI0 to handle pre-processing and post-processing steps. - Adjusted tests to utilize the renamed classes, ensuring consistency and functionality. - Enhanced clarity and maintainability by removing outdated naming conventions. * refactor(pi05): rename PI0OpenPIPolicy to PI0Policy and update configuration - Renamed `PI0OpenPIPolicy` to `PI0Policy` for consistency with naming conventions. - Updated the `PI05OpenPIConfig` to include a new `tokenizer_max_length` attribute and changed the normalization mode for state from `MEAN_STD` to `QUANTILES`. - Simplified model initialization in `PI05OpenPIPolicy` by removing unused `dataset_stats` parameter. - Added a new processor class for `Pi05PrepareStateTokenizerProcessorStep` with `@dataclass` for improved readability. - Introduced a test script to compare the integration of the PI0OpenPI policy with the original implementation, ensuring local testing compatibility. * refactor(pi05): update imports and rename configuration classes - Changed imports to reflect the new naming convention for PI05 configuration and policy classes. - Renamed `PI05OpenPIConfig` to `PI05Config` and `PI05OpenPIPolicy` to `PI05Policy` for consistency. - Introduced a new processor file for PI05, implementing pre-processing and post-processing steps. - Updated tests to utilize the renamed classes, ensuring functionality and consistency across the codebase. * update(pi05): increase tokenizer_max_length for improved processing - Changed the `tokenizer_max_length` from 48 to 200 to enhance the model's capability in handling longer sequences. - This adjustment aims to improve the overall performance and flexibility of the PI05 configuration. * add default for state (max_state_dim) * correct naming * fix import * cleanup code * remove unused test * us quantiles for action * move to device * remove discrete state assert * fix pi05 test * move pi05 to device * use base models in comparison tests * small renames for tests * change number of tokens pi05 test * fix openpi tokenization in test * fix hub test * fix test * assert lerobot vs openpi tests --------- Co-authored-by: Pepijn <pepijn@huggingface.co> * add headers * add back previously removed imports * update if statement load processor with dataset stats * remove to avoid circular import * inject dataset stats for pretrained models * check normalization before applying * add link to quantile augument script * fix(policies): transformers import for ci in PI0 & PI05 (#2039) * fix(policies): transformers import for ci in PI0 * fix(policies): transformers import for ci in PI05 * test(processor): fix expected raise when normalization types are missing (#2040) * switch normalization order pipeline for pi05 * Fix/quantiles script (#2064) * refactor augment stats with quantiles script add parallelization for faster processing shift the quantile normalization between -1 1 * fix replay buffer tests * fix comment * overwrite the pipeline normalization features with the policy features * remove double normalization overwrite * cleanup from pretrained * remove typo * also set norm_map * fix(augment_quantiles) images incorrectly divided by 255 * clamp quantiles * link to lerobot base models * rename tests * encorperate PR feedback * update docstring for RunningQuantileStats * update doc links * Revert "clamp quantiles" This reverts commit 172207471c8f2cb62958e9a9e6a0535ba3ff67d4. * fix self.paligemma * fix tests related to quantiles that were scaled to [0,1], the new range is [-1, 1] * fix libero doc and use different transformer branch * use fix branch instead of feat * update results libero * add new line * fix formatting * precommit * update results libero * update libero doc * update title * final changes * add quantiles to test * run pre commit --------- Signed-off-by: Steven Palma <imstevenpmwork@ieee.org> Co-authored-by: Michel Aractingi <michel.aractingi@huggingface.co> Co-authored-by: Adil Zouitine <adilzouitinegm@gmail.com> Co-authored-by: Steven Palma <imstevenpmwork@ieee.org> Co-authored-by: Steven Palma <steven.palma@huggingface.co>
This commit is contained in:
Pepijn
@@ -166,6 +166,226 @@ def test_min_max_normalization(observation_normalizer):
|
||||
assert torch.allclose(normalized_obs[OBS_STATE], expected_state, atol=1e-6)
|
||||
|
||||
|
||||
def test_quantile_normalization():
|
||||
"""Test QUANTILES mode using 1st-99th percentiles."""
|
||||
features = {
|
||||
"observation.state": PolicyFeature(FeatureType.STATE, (2,)),
|
||||
}
|
||||
norm_map = {
|
||||
FeatureType.STATE: NormalizationMode.QUANTILES,
|
||||
}
|
||||
stats = {
|
||||
"observation.state": {
|
||||
"q01": np.array([0.1, -0.8]), # 1st percentile
|
||||
"q99": np.array([0.9, 0.8]), # 99th percentile
|
||||
},
|
||||
}
|
||||
|
||||
normalizer = NormalizerProcessorStep(features=features, norm_map=norm_map, stats=stats)
|
||||
|
||||
observation = {
|
||||
"observation.state": torch.tensor([0.5, 0.0]),
|
||||
}
|
||||
transition = create_transition(observation=observation)
|
||||
|
||||
normalized_transition = normalizer(transition)
|
||||
normalized_obs = normalized_transition[TransitionKey.OBSERVATION]
|
||||
|
||||
# Check quantile normalization to [-1, 1]
|
||||
# For state[0]: 2 * (0.5 - 0.1) / (0.9 - 0.1) - 1 = 2 * 0.4 / 0.8 - 1 = 0.0
|
||||
# For state[1]: 2 * (0.0 - (-0.8)) / (0.8 - (-0.8)) - 1 = 2 * 0.8 / 1.6 - 1 = 0.0
|
||||
expected_state = torch.tensor([0.0, 0.0])
|
||||
assert torch.allclose(normalized_obs["observation.state"], expected_state, atol=1e-6)
|
||||
|
||||
|
||||
def test_quantile10_normalization():
|
||||
"""Test QUANTILE10 mode using 10th-90th percentiles."""
|
||||
features = {
|
||||
"observation.state": PolicyFeature(FeatureType.STATE, (2,)),
|
||||
}
|
||||
norm_map = {
|
||||
FeatureType.STATE: NormalizationMode.QUANTILE10,
|
||||
}
|
||||
stats = {
|
||||
"observation.state": {
|
||||
"q10": np.array([0.2, -0.6]), # 10th percentile
|
||||
"q90": np.array([0.8, 0.6]), # 90th percentile
|
||||
},
|
||||
}
|
||||
|
||||
normalizer = NormalizerProcessorStep(features=features, norm_map=norm_map, stats=stats)
|
||||
|
||||
observation = {
|
||||
"observation.state": torch.tensor([0.5, 0.0]),
|
||||
}
|
||||
transition = create_transition(observation=observation)
|
||||
|
||||
normalized_transition = normalizer(transition)
|
||||
normalized_obs = normalized_transition[TransitionKey.OBSERVATION]
|
||||
|
||||
# Check quantile normalization to [-1, 1]
|
||||
# For state[0]: 2 * (0.5 - 0.2) / (0.8 - 0.2) - 1 = 2 * 0.3 / 0.6 - 1 = 0.0
|
||||
# For state[1]: 2 * (0.0 - (-0.6)) / (0.6 - (-0.6)) - 1 = 2 * 0.6 / 1.2 - 1 = 0.0
|
||||
expected_state = torch.tensor([0.0, 0.0])
|
||||
assert torch.allclose(normalized_obs["observation.state"], expected_state, atol=1e-6)
|
||||
|
||||
|
||||
def test_quantile_unnormalization():
|
||||
"""Test that quantile normalization can be reversed properly."""
|
||||
features = {
|
||||
"action": PolicyFeature(FeatureType.ACTION, (2,)),
|
||||
}
|
||||
norm_map = {
|
||||
FeatureType.ACTION: NormalizationMode.QUANTILES,
|
||||
}
|
||||
stats = {
|
||||
"action": {
|
||||
"q01": np.array([0.1, -0.8]),
|
||||
"q99": np.array([0.9, 0.8]),
|
||||
},
|
||||
}
|
||||
|
||||
normalizer = NormalizerProcessorStep(features=features, norm_map=norm_map, stats=stats)
|
||||
unnormalizer = UnnormalizerProcessorStep(features=features, norm_map=norm_map, stats=stats)
|
||||
|
||||
# Test round-trip normalization
|
||||
original_action = torch.tensor([0.5, 0.0])
|
||||
transition = create_transition(action=original_action)
|
||||
|
||||
# Normalize then unnormalize
|
||||
normalized = normalizer(transition)
|
||||
unnormalized = unnormalizer(normalized)
|
||||
|
||||
# Should recover original values
|
||||
recovered_action = unnormalized[TransitionKey.ACTION]
|
||||
assert torch.allclose(recovered_action, original_action, atol=1e-6)
|
||||
|
||||
|
||||
def test_quantile_division_by_zero():
|
||||
"""Test quantile normalization handles edge case where q01 == q99."""
|
||||
features = {
|
||||
"observation.state": PolicyFeature(FeatureType.STATE, (1,)),
|
||||
}
|
||||
norm_map = {
|
||||
FeatureType.STATE: NormalizationMode.QUANTILES,
|
||||
}
|
||||
stats = {
|
||||
"observation.state": {
|
||||
"q01": np.array([0.5]), # Same value
|
||||
"q99": np.array([0.5]), # Same value -> division by zero case
|
||||
},
|
||||
}
|
||||
|
||||
normalizer = NormalizerProcessorStep(features=features, norm_map=norm_map, stats=stats)
|
||||
|
||||
observation = {
|
||||
"observation.state": torch.tensor([0.5]),
|
||||
}
|
||||
transition = create_transition(observation=observation)
|
||||
|
||||
# Should not crash and should handle gracefully
|
||||
normalized_transition = normalizer(transition)
|
||||
normalized_obs = normalized_transition[TransitionKey.OBSERVATION]
|
||||
|
||||
# When quantiles are identical, should normalize to 0 (due to epsilon handling)
|
||||
assert torch.isfinite(normalized_obs["observation.state"]).all()
|
||||
|
||||
|
||||
def test_quantile_partial_stats():
|
||||
"""Test that quantile normalization handles missing quantile stats by raising."""
|
||||
features = {
|
||||
"observation.state": PolicyFeature(FeatureType.STATE, (2,)),
|
||||
}
|
||||
norm_map = {
|
||||
FeatureType.STATE: NormalizationMode.QUANTILES,
|
||||
}
|
||||
|
||||
# Missing q99 - should pass through unchanged
|
||||
stats_partial = {
|
||||
"observation.state": {
|
||||
"q01": np.array([0.1, -0.8]), # Only q01, missing q99
|
||||
},
|
||||
}
|
||||
|
||||
normalizer = NormalizerProcessorStep(features=features, norm_map=norm_map, stats=stats_partial)
|
||||
|
||||
observation = {
|
||||
"observation.state": torch.tensor([0.5, 0.0]),
|
||||
}
|
||||
transition = create_transition(observation=observation)
|
||||
|
||||
with pytest.raises(ValueError, match="QUANTILES normalization mode requires q01 and q99 stats"):
|
||||
_ = normalizer(transition)
|
||||
|
||||
|
||||
def test_quantile_mixed_with_other_modes():
|
||||
"""Test quantile normalization mixed with other normalization modes."""
|
||||
features = {
|
||||
"observation.image": PolicyFeature(FeatureType.VISUAL, (3,)),
|
||||
"observation.state": PolicyFeature(FeatureType.STATE, (2,)),
|
||||
"action": PolicyFeature(FeatureType.ACTION, (2,)),
|
||||
}
|
||||
norm_map = {
|
||||
FeatureType.VISUAL: NormalizationMode.MEAN_STD, # Standard normalization
|
||||
FeatureType.STATE: NormalizationMode.QUANTILES, # Quantile normalization
|
||||
FeatureType.ACTION: NormalizationMode.QUANTILE10, # Different quantile mode
|
||||
}
|
||||
stats = {
|
||||
"observation.image": {"mean": [0.5, 0.5, 0.5], "std": [0.2, 0.2, 0.2]},
|
||||
"observation.state": {"q01": [0.1, -0.8], "q99": [0.9, 0.8]},
|
||||
"action": {"q10": [0.2, -0.6], "q90": [0.8, 0.6]},
|
||||
}
|
||||
|
||||
normalizer = NormalizerProcessorStep(features=features, norm_map=norm_map, stats=stats)
|
||||
|
||||
observation = {
|
||||
"observation.image": torch.tensor([0.7, 0.5, 0.3]),
|
||||
"observation.state": torch.tensor([0.5, 0.0]), # Should use QUANTILES
|
||||
}
|
||||
action = torch.tensor([0.5, 0.0]) # Should use QUANTILE10
|
||||
transition = create_transition(observation=observation, action=action)
|
||||
|
||||
normalized_transition = normalizer(transition)
|
||||
normalized_obs = normalized_transition[TransitionKey.OBSERVATION]
|
||||
normalized_action = normalized_transition[TransitionKey.ACTION]
|
||||
|
||||
# Image should be mean/std normalized: (0.7 - 0.5) / 0.2 = 1.0, etc.
|
||||
expected_image = (torch.tensor([0.7, 0.5, 0.3]) - 0.5) / 0.2
|
||||
assert torch.allclose(normalized_obs["observation.image"], expected_image)
|
||||
|
||||
# State should be quantile normalized: 2 * (0.5 - 0.1) / (0.9 - 0.1) - 1 = 0.0, etc.
|
||||
expected_state = torch.tensor([0.0, 0.0])
|
||||
assert torch.allclose(normalized_obs["observation.state"], expected_state, atol=1e-6)
|
||||
|
||||
# Action should be quantile10 normalized: 2 * (0.5 - 0.2) / (0.8 - 0.2) - 1 = 0.0, etc.
|
||||
expected_action = torch.tensor([0.0, 0.0])
|
||||
assert torch.allclose(normalized_action, expected_action, atol=1e-6)
|
||||
|
||||
|
||||
def test_quantile_with_missing_stats():
|
||||
"""Test that quantile normalization handles completely missing stats gracefully."""
|
||||
features = {
|
||||
"observation.state": PolicyFeature(FeatureType.STATE, (2,)),
|
||||
}
|
||||
norm_map = {
|
||||
FeatureType.STATE: NormalizationMode.QUANTILES,
|
||||
}
|
||||
stats = {} # No stats provided
|
||||
|
||||
normalizer = NormalizerProcessorStep(features=features, norm_map=norm_map, stats=stats)
|
||||
|
||||
observation = {
|
||||
"observation.state": torch.tensor([0.5, 0.0]),
|
||||
}
|
||||
transition = create_transition(observation=observation)
|
||||
|
||||
normalized_transition = normalizer(transition)
|
||||
normalized_obs = normalized_transition[TransitionKey.OBSERVATION]
|
||||
|
||||
# Should pass through unchanged when no stats available
|
||||
assert torch.allclose(normalized_obs["observation.state"], observation["observation.state"])
|
||||
|
||||
|
||||
def test_selective_normalization(observation_stats):
|
||||
features = _create_observation_features()
|
||||
norm_map = _create_observation_norm_map()
|
||||
@@ -547,7 +767,7 @@ def test_empty_stats():
|
||||
|
||||
|
||||
def test_partial_stats():
|
||||
"""If statistics are incomplete, the value should pass through unchanged."""
|
||||
"""If statistics are incomplete, we should raise."""
|
||||
stats = {OBS_IMAGE: {"mean": [0.5]}} # Missing std / (min,max)
|
||||
features = {OBS_IMAGE: PolicyFeature(FeatureType.VISUAL, (3, 96, 96))}
|
||||
norm_map = {FeatureType.VISUAL: NormalizationMode.MEAN_STD}
|
||||
@@ -555,8 +775,8 @@ def test_partial_stats():
|
||||
observation = {OBS_IMAGE: torch.tensor([0.7])}
|
||||
transition = create_transition(observation=observation)
|
||||
|
||||
processed = normalizer(transition)[TransitionKey.OBSERVATION]
|
||||
assert torch.allclose(processed[OBS_IMAGE], observation[OBS_IMAGE])
|
||||
with pytest.raises(ValueError, match="MEAN_STD normalization mode requires mean and std stats"):
|
||||
_ = normalizer(transition)[TransitionKey.OBSERVATION]
|
||||
|
||||
|
||||
def test_missing_action_stats_no_error():
|
||||
|
||||
@@ -1,424 +0,0 @@
|
||||
#!/usr/bin/env python
|
||||
|
||||
# Copyright 2025 The HuggingFace Inc. team. All rights reserved.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
"""Tests for PI0 policy processor."""
|
||||
|
||||
from unittest.mock import patch
|
||||
|
||||
import pytest
|
||||
import torch
|
||||
|
||||
from lerobot.configs.types import FeatureType, NormalizationMode, PolicyFeature
|
||||
from lerobot.policies.pi0.configuration_pi0 import PI0Config
|
||||
from lerobot.policies.pi0.processor_pi0 import Pi0NewLineProcessor, make_pi0_pre_post_processors
|
||||
from lerobot.processor import (
|
||||
AddBatchDimensionProcessorStep,
|
||||
DeviceProcessorStep,
|
||||
EnvTransition,
|
||||
NormalizerProcessorStep,
|
||||
ProcessorStep,
|
||||
RenameObservationsProcessorStep,
|
||||
TransitionKey,
|
||||
UnnormalizerProcessorStep,
|
||||
)
|
||||
from lerobot.processor.converters import create_transition, transition_to_batch
|
||||
from lerobot.utils.constants import ACTION, OBS_IMAGE, OBS_STATE
|
||||
|
||||
|
||||
class MockTokenizerProcessorStep(ProcessorStep):
|
||||
"""Mock tokenizer processor step for testing."""
|
||||
|
||||
def __init__(self, *args, **kwargs):
|
||||
# Accept any arguments to mimic the real TokenizerProcessorStep interface
|
||||
pass
|
||||
|
||||
def __call__(self, transition: EnvTransition) -> EnvTransition:
|
||||
# Pass through transition unchanged
|
||||
return transition
|
||||
|
||||
def transform_features(self, features):
|
||||
# Pass through features unchanged
|
||||
return features
|
||||
|
||||
|
||||
def create_default_config():
|
||||
"""Create a default PI0 configuration for testing."""
|
||||
config = PI0Config()
|
||||
config.input_features = {
|
||||
OBS_STATE: PolicyFeature(type=FeatureType.STATE, shape=(10,)),
|
||||
OBS_IMAGE: PolicyFeature(type=FeatureType.VISUAL, shape=(3, 224, 224)),
|
||||
}
|
||||
config.output_features = {
|
||||
ACTION: PolicyFeature(type=FeatureType.ACTION, shape=(6,)),
|
||||
}
|
||||
config.normalization_mapping = {
|
||||
FeatureType.STATE: NormalizationMode.MEAN_STD,
|
||||
FeatureType.VISUAL: NormalizationMode.IDENTITY,
|
||||
FeatureType.ACTION: NormalizationMode.MIN_MAX,
|
||||
}
|
||||
config.device = "cpu"
|
||||
config.tokenizer_max_length = 128
|
||||
return config
|
||||
|
||||
|
||||
def create_default_stats():
|
||||
"""Create default dataset statistics for testing."""
|
||||
return {
|
||||
OBS_STATE: {"mean": torch.zeros(10), "std": torch.ones(10)},
|
||||
OBS_IMAGE: {}, # No normalization for images
|
||||
ACTION: {"min": torch.full((6,), -1.0), "max": torch.ones(6)},
|
||||
}
|
||||
|
||||
|
||||
def test_make_pi0_processor_basic():
|
||||
"""Test basic creation of PI0 processor."""
|
||||
config = create_default_config()
|
||||
stats = create_default_stats()
|
||||
|
||||
with patch("lerobot.policies.pi0.processor_pi0.TokenizerProcessorStep", MockTokenizerProcessorStep):
|
||||
preprocessor, postprocessor = make_pi0_pre_post_processors(
|
||||
config,
|
||||
stats,
|
||||
)
|
||||
|
||||
# Check processor names
|
||||
assert preprocessor.name == "policy_preprocessor"
|
||||
assert postprocessor.name == "policy_postprocessor"
|
||||
|
||||
# Check steps in preprocessor
|
||||
assert len(preprocessor.steps) == 6
|
||||
assert isinstance(preprocessor.steps[0], RenameObservationsProcessorStep)
|
||||
assert isinstance(preprocessor.steps[1], AddBatchDimensionProcessorStep)
|
||||
assert isinstance(preprocessor.steps[2], Pi0NewLineProcessor)
|
||||
# Step 3 would be TokenizerProcessorStep but it's mocked
|
||||
assert isinstance(preprocessor.steps[4], DeviceProcessorStep)
|
||||
assert isinstance(preprocessor.steps[5], NormalizerProcessorStep)
|
||||
|
||||
# Check steps in postprocessor
|
||||
assert len(postprocessor.steps) == 2
|
||||
assert isinstance(postprocessor.steps[0], UnnormalizerProcessorStep)
|
||||
assert isinstance(postprocessor.steps[1], DeviceProcessorStep)
|
||||
|
||||
|
||||
def test_pi0_newline_processor_single_task():
|
||||
"""Test Pi0NewLineProcessor with single task string."""
|
||||
processor = Pi0NewLineProcessor()
|
||||
|
||||
# Test with task that doesn't have newline
|
||||
transition = create_transition(complementary_data={"task": "test task"})
|
||||
result = processor(transition)
|
||||
assert result[TransitionKey.COMPLEMENTARY_DATA]["task"] == "test task\n"
|
||||
|
||||
# Test with task that already has newline
|
||||
transition = create_transition(complementary_data={"task": "test task\n"})
|
||||
result = processor(transition)
|
||||
assert result[TransitionKey.COMPLEMENTARY_DATA]["task"] == "test task\n"
|
||||
|
||||
|
||||
def test_pi0_newline_processor_list_of_tasks():
|
||||
"""Test Pi0NewLineProcessor with list of task strings."""
|
||||
processor = Pi0NewLineProcessor()
|
||||
|
||||
# Test with list of tasks
|
||||
tasks = ["task1", "task2\n", "task3"]
|
||||
transition = create_transition(complementary_data={"task": tasks})
|
||||
result = processor(transition)
|
||||
expected = ["task1\n", "task2\n", "task3\n"]
|
||||
assert result[TransitionKey.COMPLEMENTARY_DATA]["task"] == expected
|
||||
|
||||
|
||||
def test_pi0_newline_processor_empty_transition():
|
||||
"""Test Pi0NewLineProcessor with empty transition."""
|
||||
processor = Pi0NewLineProcessor()
|
||||
|
||||
# Test with no complementary_data
|
||||
transition = create_transition()
|
||||
result = processor(transition)
|
||||
assert result == transition
|
||||
|
||||
# Test with complementary_data but no task
|
||||
transition = create_transition(complementary_data={"other": "data"})
|
||||
result = processor(transition)
|
||||
assert result == transition
|
||||
|
||||
# Test with None task
|
||||
transition = create_transition(complementary_data={"task": None})
|
||||
result = processor(transition)
|
||||
assert result == transition
|
||||
|
||||
|
||||
@pytest.mark.skipif(not torch.cuda.is_available(), reason="CUDA not available")
|
||||
def test_pi0_processor_cuda():
|
||||
"""Test PI0 processor with CUDA device."""
|
||||
config = create_default_config()
|
||||
config.device = "cuda"
|
||||
stats = create_default_stats()
|
||||
|
||||
# Mock the tokenizer processor to act as pass-through
|
||||
class MockTokenizerProcessorStep(ProcessorStep):
|
||||
def __init__(self, *args, **kwargs):
|
||||
pass
|
||||
|
||||
def __call__(self, transition):
|
||||
return transition
|
||||
|
||||
def state_dict(self):
|
||||
return {}
|
||||
|
||||
def load_state_dict(self, state):
|
||||
pass
|
||||
|
||||
def reset(self):
|
||||
pass
|
||||
|
||||
def get_config(self):
|
||||
return {"tokenizer_name": "google/paligemma-3b-pt-224"}
|
||||
|
||||
def transform_features(self, features):
|
||||
return features
|
||||
|
||||
with patch("lerobot.policies.pi0.processor_pi0.TokenizerProcessorStep", MockTokenizerProcessorStep):
|
||||
preprocessor, postprocessor = make_pi0_pre_post_processors(
|
||||
config,
|
||||
stats,
|
||||
)
|
||||
|
||||
# Create CPU data
|
||||
observation = {
|
||||
OBS_STATE: torch.randn(10),
|
||||
OBS_IMAGE: torch.randn(3, 224, 224),
|
||||
}
|
||||
action = torch.randn(6)
|
||||
transition = create_transition(observation, action, complementary_data={"task": "test task"})
|
||||
batch = transition_to_batch(transition)
|
||||
|
||||
# Process through preprocessor
|
||||
processed = preprocessor(batch)
|
||||
|
||||
# Check that data is on CUDA
|
||||
assert processed[OBS_STATE].device.type == "cuda"
|
||||
assert processed[OBS_IMAGE].device.type == "cuda"
|
||||
assert processed[TransitionKey.ACTION.value].device.type == "cuda"
|
||||
|
||||
|
||||
@pytest.mark.skipif(not torch.cuda.is_available(), reason="CUDA not available")
|
||||
def test_pi0_processor_accelerate_scenario():
|
||||
"""Test PI0 processor in simulated Accelerate scenario."""
|
||||
config = create_default_config()
|
||||
config.device = "cuda:0"
|
||||
stats = create_default_stats()
|
||||
|
||||
# Mock the tokenizer processor to act as pass-through
|
||||
class MockTokenizerProcessorStep(ProcessorStep):
|
||||
def __init__(self, *args, **kwargs):
|
||||
pass
|
||||
|
||||
def __call__(self, transition):
|
||||
return transition
|
||||
|
||||
def state_dict(self):
|
||||
return {}
|
||||
|
||||
def load_state_dict(self, state):
|
||||
pass
|
||||
|
||||
def reset(self):
|
||||
pass
|
||||
|
||||
def get_config(self):
|
||||
return {"tokenizer_name": "google/paligemma-3b-pt-224"}
|
||||
|
||||
def transform_features(self, features):
|
||||
return features
|
||||
|
||||
with patch("lerobot.policies.pi0.processor_pi0.TokenizerProcessorStep", MockTokenizerProcessorStep):
|
||||
preprocessor, postprocessor = make_pi0_pre_post_processors(
|
||||
config,
|
||||
stats,
|
||||
)
|
||||
|
||||
# Simulate Accelerate: data already on GPU and batched
|
||||
device = torch.device("cuda:0")
|
||||
observation = {
|
||||
OBS_STATE: torch.randn(1, 10).to(device),
|
||||
OBS_IMAGE: torch.randn(1, 3, 224, 224).to(device),
|
||||
}
|
||||
action = torch.randn(1, 6).to(device)
|
||||
transition = create_transition(observation, action, complementary_data={"task": ["test task"]})
|
||||
batch = transition_to_batch(transition)
|
||||
|
||||
# Process through preprocessor
|
||||
processed = preprocessor(batch)
|
||||
|
||||
# Check that data stays on same GPU
|
||||
assert processed[OBS_STATE].device == device
|
||||
assert processed[OBS_IMAGE].device == device
|
||||
assert processed[TransitionKey.ACTION.value].device == device
|
||||
|
||||
|
||||
@pytest.mark.skipif(torch.cuda.device_count() < 2, reason="Requires at least 2 GPUs")
|
||||
def test_pi0_processor_multi_gpu():
|
||||
"""Test PI0 processor with multi-GPU setup."""
|
||||
config = create_default_config()
|
||||
config.device = "cuda:0"
|
||||
stats = create_default_stats()
|
||||
|
||||
# Mock the tokenizer processor to act as pass-through
|
||||
class MockTokenizerProcessorStep(ProcessorStep):
|
||||
def __init__(self, *args, **kwargs):
|
||||
pass
|
||||
|
||||
def __call__(self, transition):
|
||||
return transition
|
||||
|
||||
def state_dict(self):
|
||||
return {}
|
||||
|
||||
def load_state_dict(self, state):
|
||||
pass
|
||||
|
||||
def reset(self):
|
||||
pass
|
||||
|
||||
def get_config(self):
|
||||
return {"tokenizer_name": "google/paligemma-3b-pt-224"}
|
||||
|
||||
def transform_features(self, features):
|
||||
return features
|
||||
|
||||
with patch("lerobot.policies.pi0.processor_pi0.TokenizerProcessorStep", MockTokenizerProcessorStep):
|
||||
preprocessor, postprocessor = make_pi0_pre_post_processors(
|
||||
config,
|
||||
stats,
|
||||
)
|
||||
|
||||
# Simulate data on different GPU
|
||||
device = torch.device("cuda:1")
|
||||
observation = {
|
||||
OBS_STATE: torch.randn(1, 10).to(device),
|
||||
OBS_IMAGE: torch.randn(1, 3, 224, 224).to(device),
|
||||
}
|
||||
action = torch.randn(1, 6).to(device)
|
||||
transition = create_transition(observation, action, complementary_data={"task": ["test task"]})
|
||||
batch = transition_to_batch(transition)
|
||||
|
||||
# Process through preprocessor
|
||||
processed = preprocessor(batch)
|
||||
|
||||
# Check that data stays on cuda:1
|
||||
assert processed[OBS_STATE].device == device
|
||||
assert processed[OBS_IMAGE].device == device
|
||||
assert processed[TransitionKey.ACTION.value].device == device
|
||||
|
||||
|
||||
def test_pi0_processor_without_stats():
|
||||
"""Test PI0 processor creation without dataset statistics."""
|
||||
config = create_default_config()
|
||||
|
||||
# Mock the tokenizer processor
|
||||
with patch("lerobot.policies.pi0.processor_pi0.TokenizerProcessorStep", MockTokenizerProcessorStep):
|
||||
preprocessor, postprocessor = make_pi0_pre_post_processors(
|
||||
config,
|
||||
dataset_stats=None,
|
||||
)
|
||||
|
||||
# Should still create processors
|
||||
assert preprocessor is not None
|
||||
assert postprocessor is not None
|
||||
|
||||
|
||||
def test_pi0_newline_processor_state_dict():
|
||||
"""Test Pi0NewLineProcessor state dict methods."""
|
||||
processor = Pi0NewLineProcessor()
|
||||
|
||||
# Test state_dict (should be empty)
|
||||
state = processor.state_dict()
|
||||
assert state == {}
|
||||
|
||||
# Test load_state_dict (should do nothing)
|
||||
processor.load_state_dict({})
|
||||
|
||||
# Test reset (should do nothing)
|
||||
processor.reset()
|
||||
|
||||
# Test get_config
|
||||
config = processor.get_config()
|
||||
assert config == {}
|
||||
|
||||
|
||||
@pytest.mark.skipif(not torch.cuda.is_available(), reason="CUDA not available")
|
||||
def test_pi0_processor_bfloat16_device_float32_normalizer():
|
||||
"""Test: DeviceProcessor(bfloat16) + NormalizerProcessor(float32) → output bfloat16 via automatic adaptation"""
|
||||
config = create_default_config()
|
||||
stats = create_default_stats()
|
||||
config.device = "cuda"
|
||||
|
||||
with patch("lerobot.policies.pi0.processor_pi0.TokenizerProcessorStep", MockTokenizerProcessorStep):
|
||||
preprocessor, _ = make_pi0_pre_post_processors(
|
||||
config,
|
||||
stats,
|
||||
)
|
||||
|
||||
# Modify the pipeline to use bfloat16 device processor with float32 normalizer
|
||||
modified_steps = []
|
||||
for step in preprocessor.steps:
|
||||
if isinstance(step, DeviceProcessorStep):
|
||||
# Device processor converts to bfloat16
|
||||
modified_steps.append(DeviceProcessorStep(device=config.device, float_dtype="bfloat16"))
|
||||
elif isinstance(step, NormalizerProcessorStep):
|
||||
# Normalizer stays configured as float32 (will auto-adapt to bfloat16)
|
||||
norm_step = step # Now type checker knows this is NormalizerProcessorStep
|
||||
modified_steps.append(
|
||||
NormalizerProcessorStep(
|
||||
features=norm_step.features,
|
||||
norm_map=norm_step.norm_map,
|
||||
stats=norm_step.stats,
|
||||
device=config.device,
|
||||
dtype=torch.float32, # Deliberately configured as float32
|
||||
)
|
||||
)
|
||||
else:
|
||||
modified_steps.append(step)
|
||||
preprocessor.steps = modified_steps
|
||||
|
||||
# Verify initial normalizer configuration (PI0 has NormalizerProcessorStep at index 5)
|
||||
normalizer_step = preprocessor.steps[5] # NormalizerProcessorStep
|
||||
assert normalizer_step.dtype == torch.float32
|
||||
|
||||
# Create test data with both state and visual observations
|
||||
observation = {
|
||||
OBS_STATE: torch.randn(10, dtype=torch.float32), # PI0 expects size 10
|
||||
OBS_IMAGE: torch.randn(3, 224, 224, dtype=torch.float32),
|
||||
}
|
||||
action = torch.randn(6, dtype=torch.float32) # PI0 expects size 6
|
||||
transition = create_transition(
|
||||
observation, action, complementary_data={"task": "test bfloat16 adaptation"}
|
||||
)
|
||||
batch = transition_to_batch(transition)
|
||||
|
||||
# Process through full pipeline
|
||||
processed = preprocessor(batch)
|
||||
|
||||
# Verify: DeviceProcessor → bfloat16, NormalizerProcessor adapts → final output is bfloat16
|
||||
assert processed[OBS_STATE].dtype == torch.bfloat16
|
||||
assert processed[OBS_IMAGE].dtype == torch.bfloat16 # IDENTITY normalization still gets dtype conversion
|
||||
assert processed[TransitionKey.ACTION.value].dtype == torch.bfloat16
|
||||
|
||||
# Verify normalizer automatically adapted its internal state
|
||||
assert normalizer_step.dtype == torch.bfloat16
|
||||
# Check state stats (has normalization)
|
||||
for stat_tensor in normalizer_step._tensor_stats[OBS_STATE].values():
|
||||
assert stat_tensor.dtype == torch.bfloat16
|
||||
# OBS_IMAGE uses IDENTITY normalization, so no stats to check
|
||||
Reference in New Issue
Block a user