The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: CastError
Message: Couldn't cast
id: string
original_payload_bytes: int64
original_files: int64
data_parts: string
source_payloads_unchanged: bool
metadata_binding: null
unpickling_performed: bool
exact_training_resume_certified: bool
payload_verification: string
dependencies: string
all_native_files_required: bool
scope: string
preparation_complete: bool
files: list<item: struct<path: string, bytes: int64, sha256: string, parts: list<item: struct<path: string, (... 283 chars omitted)
child 0, item: struct<path: string, bytes: int64, sha256: string, parts: list<item: struct<path: string, bytes: int (... 271 chars omitted)
child 0, path: string
child 1, bytes: int64
child 2, sha256: string
child 3, parts: list<item: struct<path: string, bytes: int64, sha256: string>>
child 0, item: struct<path: string, bytes: int64, sha256: string>
child 0, path: string
child 1, bytes: int64
child 2, sha256: string
child 4, metadata_review: struct<pickle_sha256: string, credential_contact_pattern_hits: int64, embedded_binding_literals: str (... 76 chars omitted)
child 0, pickle_sha256: string
child 1, credential_contact_pattern_hits: int64
child 2, embedded_binding_literals: struct<>
child 3, unpickling_performed: bool
child 4, numeric_tensor_storage_not_decoded: bool
child 5, source_unchanged_stat: bool
child 6, source_component: string
to
{'files': List({'path': Value('string'), 'bytes': Value('int64'), 'sha256': Value('string'), 'parts': List({'path': Value('string'), 'bytes': Value('int64'), 'sha256': Value('string')}), 'metadata_review': {'pickle_sha256': Value('string'), 'credential_contact_pattern_hits': Value('int64'), 'embedded_binding_literals': {}, 'unpickling_performed': Value('bool'), 'numeric_tensor_storage_not_decoded': Value('bool')}, 'source_unchanged_stat': Value('bool'), 'source_component': Value('string')}), 'preparation_complete': Value('bool')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 149, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 129, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 544, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 3052, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2559, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2584, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 558, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 440, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 361, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2381, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2309, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
id: string
original_payload_bytes: int64
original_files: int64
data_parts: string
source_payloads_unchanged: bool
metadata_binding: null
unpickling_performed: bool
exact_training_resume_certified: bool
payload_verification: string
dependencies: string
all_native_files_required: bool
scope: string
preparation_complete: bool
files: list<item: struct<path: string, bytes: int64, sha256: string, parts: list<item: struct<path: string, (... 283 chars omitted)
child 0, item: struct<path: string, bytes: int64, sha256: string, parts: list<item: struct<path: string, bytes: int (... 271 chars omitted)
child 0, path: string
child 1, bytes: int64
child 2, sha256: string
child 3, parts: list<item: struct<path: string, bytes: int64, sha256: string>>
child 0, item: struct<path: string, bytes: int64, sha256: string>
child 0, path: string
child 1, bytes: int64
child 2, sha256: string
child 4, metadata_review: struct<pickle_sha256: string, credential_contact_pattern_hits: int64, embedded_binding_literals: str (... 76 chars omitted)
child 0, pickle_sha256: string
child 1, credential_contact_pattern_hits: int64
child 2, embedded_binding_literals: struct<>
child 3, unpickling_performed: bool
child 4, numeric_tensor_storage_not_decoded: bool
child 5, source_unchanged_stat: bool
child 6, source_component: string
to
{'files': List({'path': Value('string'), 'bytes': Value('int64'), 'sha256': Value('string'), 'parts': List({'path': Value('string'), 'bytes': Value('int64'), 'sha256': Value('string')}), 'metadata_review': {'pickle_sha256': Value('string'), 'credential_contact_pattern_hits': Value('int64'), 'embedded_binding_literals': {}, 'unpickling_performed': Value('bool'), 'numeric_tensor_storage_not_decoded': Value('bool')}, 'source_unchanged_stat': Value('bool'), 'source_component': Value('string')}), 'preparation_complete': Value('bool')}
because column names don't matchNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
GeoEvidence vision step23 complete native FSDP checkpoint
Native checkpoint payload: 22633459030 bytes / 19 original files, transported in <=64MiB parts. All19 native files are included: six model shards, six optimizer shards, six extra_state files containing RNG/scheduler state, and data.pt data-cursor state. These are same-world-size6 native training-state components, not merely six extra-state sidecars. Existing fixed-revision HF weights/config/tokenizer dependencies are mapped in dependencies and are not retransmitted as a new inference export.
Source metadata was inspected using ZIP directory and pickle opcode parsing without unpickling. Credential/contact patterns were checked in textual metadata; numeric tensor storage was not interpreted as content. Original checkpoint files were streamed to parts with SHA256 and unchanged-source stat checks. This archive operation did not load or execute the pickle, launch training, or verify a fresh GPU exact-resume trajectory.
Dependency/environment gaps remain explicit: historical installed package versions and exact runtime source checkout are not fully certified. Source-reference files are static .py.txt, not an instruction to run historical scripts; some contain checkpoint pruning or training paths. Original safe research archival authorization does not authorize destructive restore operations. The manifest may mention parts still uploading until the external publication receipt or final expected-file validation reports committed_verified.
Recover bytes
hf download wangzyuan/GeoEvidence-Vision-Step23-Native-Checkpoint-Rescue --repo-type dataset --revision COMMIT --local-dir ./checkpoint-rescue
python ./checkpoint-rescue/restore_research_data.py ./checkpoint-rescue
The restorer hashes every local part and complete output and publishes no output on corruption. It never unpickles. Afterwards use the original trusted runtime and matched pinned dependencies under separate review. Do not blindly torch.load untrusted files or infer exact continuation from mere byte recovery. Full-payload remote verification for this large archive uses committed fixed-revision LFS SHA256/size metadata; supporting files are actually anonymously downloaded. A fresh full checkpoint download/reconstruction or training resume is not claimed.
- Downloads last month
- 114