admin/lerobot
1
0
Fork 0
mirror of https://github.com/huggingface/lerobot.git synced 2026-05-26 14:09:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

refactor(dataset): split LeRobotDataset into DatasetReader & DatasetWriter (+ API cleanup) (#3180)

Browse Source 
* refactor(dataset): split reader and writer

* chore(dataset): remove proxys

* refactor(dataset): better reader & writer encapsulation

* refactor(datasets): clean API + reduce leaky implementations

* refactor(dataset): API cleaning for writer, reader and meta

* refactor(dataset): expose writer & reader + other minor improvements

* refactor(dataset): improve teardown routine

* refactor(dataset): add hf_dataset property at the facade level

* chore(dataset): add init for datasset module

* docs(dataset): add docstrings for public API of the dataset classes

* tests(dataset): add tests for new classes

* fix(dataset): remove circular dependecy
This commit is contained in:
Steven Palma
2026-03-26 19:09:25 +01:00
committed by GitHub
parent 017ff73fbf
commit 123495250b
28 changed files with 2742 additions and 1158 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/artifacts/policies/save_policy_to_safetensors.py
+1 -1
View File
@@ -80,7 +80,7 @@ def get_policy_stats(ds_repo_id: str, policy_name: str, policy_kwargs: dict):
# HACK: We reload a batch with no delta_indices as `select_action` won't expect a timestamps dimension
# We simulate having an environment using a dataset by setting delta_indices to None and dropping tensors
# indicating padding (those ending with "_is_pad")
dataset.delta_indices = None
dataset.reader.delta_indices = None
batch = next(iter(dataloader))
obs = {}
for k in batch: