Dataset Viewer
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code: StreamingRowsError
Exception: CastError
Message: Couldn't cast
engine: string
counts: struct<ok: int64>
child 0, ok: int64
successful_request_latency_ms: struct<median: double, p95: double, mean: double>
child 0, median: double
child 1, p95: double
child 2, mean: double
benchmarks: list<item: struct<catalog_id: int64, dataset: string, requests: int64, answered: int64, unsupported: (... 6317 chars omitted)
child 0, item: struct<catalog_id: int64, dataset: string, requests: int64, answered: int64, unsupported: int64, err (... 6305 chars omitted)
child 0, catalog_id: int64
child 1, dataset: string
child 2, requests: int64
child 3, answered: int64
child 4, unsupported: int64
child 5, errors: int64
child 6, abstained: int64
child 7, pending: int64
child 8, scored_requests: int64
child 9, metric: string
child 10, score: double
child 11, reference_same_cases: null
child 12, median_ms: double
child 13, detail: struct<field_accuracy: double, case_exact_accuracy: double, custom_metrics: struct<ndcg_at_10: doubl (... 5311 chars omitted)
child 0, field_accuracy: double
child 1, case_exact_accuracy: double
child 2, custom_metrics: struct<ndcg_at_10: double, mrr: double, recall_at_10: double, candidate_recall: double, scorable_can (... 452 chars omitted)
child 0, ndcg_at_10: double
child 1, mrr: double
child 2, recall_at_10: double
child 3, candidate_recall: double
...
, runs/wemm-jev-vision/environment.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
child 8, runs/wemm-jev-vision/results.jsonl: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
child 9, runs/wemm-jev-vision/status.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
child 10, runs/wemm-jev-vision/vision-scores.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
model: struct<repo_id: string, revision: string>
child 0, repo_id: string
child 1, revision: string
format: string
public_results: struct<text: struct<edition: string, public_index: double, scoreable_requests: int64, complete: bool (... 90 chars omitted)
child 0, text: struct<edition: string, public_index: double, scoreable_requests: int64, complete: bool>
child 0, edition: string
child 1, public_index: double
child 2, scoreable_requests: int64
child 3, complete: bool
child 1, vision: struct<edition: string, public_index: double, requests: int64, complete: bool>
child 0, edition: string
child 1, public_index: double
child 2, requests: int64
child 3, complete: bool
inference: struct<repo_id: string, path: string, manifest_sha256: string, revision: string>
child 0, repo_id: string
child 1, path: string
child 2, manifest_sha256: string
child 3, revision: string
to
{'format': Value('string'), 'model': {'repo_id': Value('string'), 'revision': Value('string')}, 'public_results': {'text': {'edition': Value('string'), 'public_index': Value('float64'), 'scoreable_requests': Value('int64'), 'complete': Value('bool')}, 'vision': {'edition': Value('string'), 'public_index': Value('float64'), 'requests': Value('int64'), 'complete': Value('bool')}}, 'files': {'README.md': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-text/benchmark-summary.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-text/environment.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-text/index.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-text/results.jsonl.gz': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-text/scores.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-text/status.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-vision/environment.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-vision/results.jsonl': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-vision/status.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-vision/vision-scores.json': {'bytes': Value('int64'), 'sha256': Value('string')}}, 'inference': {'repo_id': Value('string'), 'path': Value('string'), 'manifest_sha256': Value('string'), 'revision': Value('string')}}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 147, 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 127, 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 483, 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 2951, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2461, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2486, 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 547, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 430, 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 343, 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 2378, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2306, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
engine: string
counts: struct<ok: int64>
child 0, ok: int64
successful_request_latency_ms: struct<median: double, p95: double, mean: double>
child 0, median: double
child 1, p95: double
child 2, mean: double
benchmarks: list<item: struct<catalog_id: int64, dataset: string, requests: int64, answered: int64, unsupported: (... 6317 chars omitted)
child 0, item: struct<catalog_id: int64, dataset: string, requests: int64, answered: int64, unsupported: int64, err (... 6305 chars omitted)
child 0, catalog_id: int64
child 1, dataset: string
child 2, requests: int64
child 3, answered: int64
child 4, unsupported: int64
child 5, errors: int64
child 6, abstained: int64
child 7, pending: int64
child 8, scored_requests: int64
child 9, metric: string
child 10, score: double
child 11, reference_same_cases: null
child 12, median_ms: double
child 13, detail: struct<field_accuracy: double, case_exact_accuracy: double, custom_metrics: struct<ndcg_at_10: doubl (... 5311 chars omitted)
child 0, field_accuracy: double
child 1, case_exact_accuracy: double
child 2, custom_metrics: struct<ndcg_at_10: double, mrr: double, recall_at_10: double, candidate_recall: double, scorable_can (... 452 chars omitted)
child 0, ndcg_at_10: double
child 1, mrr: double
child 2, recall_at_10: double
child 3, candidate_recall: double
...
, runs/wemm-jev-vision/environment.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
child 8, runs/wemm-jev-vision/results.jsonl: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
child 9, runs/wemm-jev-vision/status.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
child 10, runs/wemm-jev-vision/vision-scores.json: struct<bytes: int64, sha256: string>
child 0, bytes: int64
child 1, sha256: string
model: struct<repo_id: string, revision: string>
child 0, repo_id: string
child 1, revision: string
format: string
public_results: struct<text: struct<edition: string, public_index: double, scoreable_requests: int64, complete: bool (... 90 chars omitted)
child 0, text: struct<edition: string, public_index: double, scoreable_requests: int64, complete: bool>
child 0, edition: string
child 1, public_index: double
child 2, scoreable_requests: int64
child 3, complete: bool
child 1, vision: struct<edition: string, public_index: double, requests: int64, complete: bool>
child 0, edition: string
child 1, public_index: double
child 2, requests: int64
child 3, complete: bool
inference: struct<repo_id: string, path: string, manifest_sha256: string, revision: string>
child 0, repo_id: string
child 1, path: string
child 2, manifest_sha256: string
child 3, revision: string
to
{'format': Value('string'), 'model': {'repo_id': Value('string'), 'revision': Value('string')}, 'public_results': {'text': {'edition': Value('string'), 'public_index': Value('float64'), 'scoreable_requests': Value('int64'), 'complete': Value('bool')}, 'vision': {'edition': Value('string'), 'public_index': Value('float64'), 'requests': Value('int64'), 'complete': Value('bool')}}, 'files': {'README.md': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-text/benchmark-summary.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-text/environment.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-text/index.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-text/results.jsonl.gz': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-text/scores.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-text/status.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-vision/environment.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-vision/results.jsonl': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-vision/status.json': {'bytes': Value('int64'), 'sha256': Value('string')}, 'runs/wemm-jev-vision/vision-scores.json': {'bytes': Value('int64'), 'sha256': Value('string')}}, 'inference': {'repo_id': Value('string'), 'path': Value('string'), 'manifest_sha256': Value('string'), 'revision': Value('string')}}
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.
WeMM-JEV Results
Decision Index results for WeMM-JEV.
| Suite | Public index | Scored requests |
|---|---|---|
| Text 0.3 | 69.37 | 140,178 |
| Vision 0.3 | 77.97 | 12,210 |
Text predictions contain 140,620 records, including 442 official exclusions.
Results and environments: Text · Vision. File sizes and SHA256 hashes: artifact_manifest.json.
Inference
Code and dependencies: model repository's inference module.
BF16 backbone, FP32 readout, SDPA, noncausal full attention, last-token pooling, temperature 1.0, batch size 8, padded-token budget 8192, no permutation averaging. Vision image cap: 1,638,400 pixels.
Run
Use Python 3.12 and CUDA 12.8. Run from this repository's root. Skip PyTorch installation if PyTorch 2.8.0 and torchvision 0.23.0 with CUDA 12.8 are already installed.
hf download Rover00/WeMM-JEV --revision c28b30fd66fc2269500b93b97bb5b79980c9ec05 --local-dir model
python -m pip install torch==2.8.0 torchvision==0.23.0 --index-url https://download.pytorch.org/whl/cu128
python -m pip install -r model/inference/requirements.txt
python -m pip install --no-build-isolation --no-deps causal-conv1d==1.7.0
python -m pip install --no-deps 'triton>=3.7.1'
python -m pip install 'decision-index[vision] @ git+https://github.com/apolinario/decision-index.git@f8552a13b643d39f61185f5951bb5fed16bedf51'
Prepare the public suites using the official instructions.
Set TEXT_SUITE and VISION_SUITE to their directories.
export PYTHONPATH="$PWD/model${PYTHONPATH:+:$PYTHONPATH}"
python -m decision_index pipeline --edition 0.3 \
--suite-dir "$TEXT_SUITE" \
--engine inference.engine:D3Engine \
--option model="$PWD/model" --option device=cuda:0 \
--option batch_size=8 --option permutation_average=0 --compact \
--out new-runs/text
python -m decision_index vision pipeline --dir "$VISION_SUITE" \
--engine inference.engine:D3Engine \
--option model="$PWD/model" --option device=cuda:0 \
--option batch_size=8 --option permutation_average=0 --workers 1 --compact \
--out new-runs/vision
Scoring kit commits: Text 9eb2dbe2a358004c8782c66e40a83ac07b953fec;
Vision f8552a13b643d39f61185f5951bb5fed16bedf51.
- Downloads last month
- 154