Dataset Viewer
The dataset viewer is not available for this subset.
Cannot get the split names for the config 'default' of the dataset.
Exception: SplitsNotFoundError
Message: The split names could not be parsed from the dataset config.
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 286, in get_dataset_config_info
for split_generator in builder._split_generators(
~~~~~~~~~~~~~~~~~~~~~~~~~^
StreamingDownloadManager(base_path=builder.base_path, download_config=download_config)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 83, in _split_generators
raise ValueError(
...<2 lines>...
)
ValueError: The TAR archives of the dataset should be in WebDataset format, but the files in the archive don't share the same prefix or the same types.
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/split_names.py", line 66, in compute_split_names_from_streaming_response
for split in get_dataset_split_names(
~~~~~~~~~~~~~~~~~~~~~~~^
path=dataset,
^^^^^^^^^^^^^
config_name=config,
^^^^^^^^^^^^^^^^^^^
token=hf_token,
^^^^^^^^^^^^^^^
)
^
File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 340, in get_dataset_split_names
info = get_dataset_config_info(
path,
...<6 lines>...
**config_kwargs,
)
File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 291, in get_dataset_config_info
raise SplitsNotFoundError("The split names could not be parsed from the dataset config.") from err
datasets.inspect.SplitsNotFoundError: The split names could not be parsed from the dataset config.Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
DA3-XVLA — Experiment Record (notes · results · logs)
Backup of the DA3-XVLA experimental record, captured before the Sines server decommission.
The code lives in JackLiu0406/da3-xvla-pipeline, the models in
JackLiu0406/DA3-XVLA-roboreal-ablations, and Xuan's vanilla backup in
JackLiu0406/xvla-vanilla-backup. This repo is the record: what we learned and what we measured.
Contents
| Path | What it is |
|---|---|
project_notes/ |
Development knowledge — bug diagnoses, fixes, architecture decisions, eval recipe, cluster gotchas (50 notes; the working memory, otherwise lost with the root fs). |
MODELS_TRAINED_SUMMARY.md |
Index of every model trained + its purpose. |
K320R8_ckpt90000_per_task_vs_xuan.md · *_matched_vs_xuan*.json |
Headline per-task SR/HSR for the K320 model vs Xuan's vanilla baseline. |
multinative_ckpt125000_per_task_SR.txt |
Per-task SR for the multi-native run. |
eval_summaries.tar.gz |
Full tree of per-model eval metric files (.md/.json/.txt) across all runs. |
eval_results_full.tar |
Complete eval_result/ — incl. rollout videos + raw per-episode outputs (~600 MB). |
training_logs.tar.gz |
Training stdout / loss curves for every run (v1_smoke/_logs). |
eval_ledger.tar.gz |
Consolidated eval ledger. |
Key findings (see the summaries for numbers)
- DA3-XVLA evaluated per-task vs Xuan's vanilla X-VLA baseline (matched seeds) on RoboReal/RoboTwin, clean + d6/d10/d15 clutter; report per-scene SR and HSR.
- The frozen-DA3 recipe outperformed unfrozen; the three spatial-expert arms (sequential / interleaved / interleaved+T5) are the final study (see the ablations repo).
- Downloads last month
- 96