Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed because of a cast error
Error code: DatasetGenerationCastError
Exception: DatasetGenerationCastError
Message: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 5 new columns ({'loops_d_actual', 'C_new', 'A_new', 'D_new', 'B_new'})
This happened while the csv dataset builder was generating data using
hf://datasets/GuangsTrip/surgbench-cot-smoketest/strict_swap_smoke_200/surgbench_d_rewrite/test/test_smoke_200_01.tsv (at revision dc8c8cf351fbc49c50969b0bb0d8a524fda7d74c), ['hf://datasets/GuangsTrip/surgbench-cot-smoketest@dc8c8cf351fbc49c50969b0bb0d8a524fda7d74c/strict_swap_smoke_200/surgbench_d_norewrite/test/test_smoke_200_01.tsv', 'hf://datasets/GuangsTrip/surgbench-cot-smoketest@dc8c8cf351fbc49c50969b0bb0d8a524fda7d74c/strict_swap_smoke_200/surgbench_d_rewrite/test/test_smoke_200_01.tsv']
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1848, in _prepare_split_single
writer.write_table(table)
~~~~~~~~~~~~~~~~~~^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 765, in write_table
self._write_table(pa_table, writer_batch_size=writer_batch_size)
~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 773, in _write_table
pa_table = table_cast(pa_table, self._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
A: string
A_new: string
B: string
B_new: string
C: string
C_new: string
D: string
D_new: string
answer: string
category: double
citations: string
dataset: double
diagnostics: string
image: string
index: int64
loops_d_actual: int64
loops_q_actual: int64
modality: double
new_question: string
original_question: string
smoke_semantic_task: string
smoke_source_dataset: string
smoke_source_tsv: string
smoke_variant: string
variants_tried: int64
-- schema metadata --
pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 3214
to
{'A': Value('string'), 'B': Value('string'), 'C': Value('string'), 'D': Value('string'), 'answer': Value('string'), 'category': Value('float64'), 'citations': Value('string'), 'dataset': Value('float64'), 'diagnostics': Value('string'), 'image': Value('string'), 'index': Value('int64'), 'loops_q_actual': Value('int64'), 'modality': Value('float64'), 'new_question': Value('string'), 'original_question': Value('string'), 'smoke_semantic_task': Value('string'), 'smoke_source_dataset': Value('string'), 'smoke_source_tsv': Value('string'), 'smoke_variant': Value('string'), 'variants_tried': Value('int64')}
because column names don't match
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1369, in compute_config_parquet_and_info_response
parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
~~~~~~~~~~~~~~~~~~~~~~~~~^
builder, max_dataset_size_bytes=max_dataset_size_bytes
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 948, in stream_convert_to_parquet
builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1694, in _prepare_split
for job_id, done, content in self._prepare_split_single(
~~~~~~~~~~~~~~~~~~~~~~~~~~^
gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
):
^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1850, in _prepare_split_single
raise DatasetGenerationCastError.from_cast_error(
...<4 lines>...
)
datasets.exceptions.DatasetGenerationCastError: An error occurred while generating the dataset
All the data files must have the same columns, but at some point there are 5 new columns ({'loops_d_actual', 'C_new', 'A_new', 'D_new', 'B_new'})
This happened while the csv dataset builder was generating data using
hf://datasets/GuangsTrip/surgbench-cot-smoketest/strict_swap_smoke_200/surgbench_d_rewrite/test/test_smoke_200_01.tsv (at revision dc8c8cf351fbc49c50969b0bb0d8a524fda7d74c), ['hf://datasets/GuangsTrip/surgbench-cot-smoketest@dc8c8cf351fbc49c50969b0bb0d8a524fda7d74c/strict_swap_smoke_200/surgbench_d_norewrite/test/test_smoke_200_01.tsv', 'hf://datasets/GuangsTrip/surgbench-cot-smoketest@dc8c8cf351fbc49c50969b0bb0d8a524fda7d74c/strict_swap_smoke_200/surgbench_d_rewrite/test/test_smoke_200_01.tsv']
Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)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.
A string | B string | C string | D string | answer string | category null | citations string | dataset null | diagnostics string | image string | index int64 | loops_q_actual int64 | modality null | new_question string | original_question string | smoke_semantic_task string | smoke_source_dataset string | smoke_source_tsv string | smoke_variant string | variants_tried int64 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
ring_curette | suction | tissue_glue | bipolar_forceps | B | null | "[{\"source\": \"PDF\", \"id\": \"kupdf.com_a-complete-hospital-manual-of-instruments-and-procedures(...TRUNCATED) | null | "{\"q_rewrite\": {\"question\": \"The picture below shows a handheld device that creates a vacuum to(...TRUNCATED) | "['/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAIBAQEBAQIBAQECAgICAgQDAgICAgUEBAMEBgUGBgYFBgYGBwkIBgcJBwYGCAsIC(...TRUNCATED) | 1,736 | 1 | null | "The picture below shows a handheld surgical instrument commonly used during procedures. Which instr(...TRUNCATED) | Identify the instrument in these images. | instrumentrecognition | pitvis | pitvis_instrumentRecognition_train_01.tsv | surgbench_d_norewrite | 1 |
Grasper | Blood | Gastrointestinal Tract | Abdominal Wall | A | null | "[{\"source\": \"PDF\", \"id\": \"Approach to Diaphragmatic Injuries.pdf:c2\", \"snippet\": \"ly the(...TRUNCATED) | null | "{\"q_rewrite\": {\"question\": \"When viewing the attached operative image, which of the following (...TRUNCATED) | "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAIBAQEBAQIBAQECAgICAgQDAgICAgUEBAMEBgUGBgYFBgYGBwkIBgcJBwYGCAsICQo(...TRUNCATED) | 297 | 0 | null | "When viewing the attached operative image, which of the following items is most distinctly outlined(...TRUNCATED) | Which object is clearly outlined in this image? | instrumentrecognition | CholecSeg8k | CholecSeg8k_InstrumentRecognition_train_01.tsv | surgbench_d_norewrite | 1 |
stapler | monopolar curved scissors | clip applier | large needle driver | B | null | "[{\"source\": \"PDF\", \"id\": \"Approach to Diaphragmatic Injuries.pdf:c9\", \"snippet\": \"tions (...TRUNCATED) | null | "{\"q_rewrite\": {\"question\": \"Identify the surgical instrument shown in the upper‑right quadra(...TRUNCATED) | "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAIBAQEBAQIBAQECAgICAgQDAgICAgUEBAMEBgUGBgYFBgYGBwkIBgcJBwYGCAsICQo(...TRUNCATED) | 510 | 0 | null | Identify the surgical instrument shown in the upper‑right quadrant of the illustration. | What is the tool in the right_top? | instrumentrecognition | endovis18 | endovis18_instrumentRecognition_train_01.tsv | surgbench_d_norewrite | 1 |
Liver Ligament | Cystic Duct | Gallbladder | L-hook Electrocautery | D | null | [] | null | "{\"error\": \"parse_failed_after_retries\", \"message\": \"question review response parse failed\"}(...TRUNCATED) | "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAIBAQEBAQIBAQECAgICAgQDAgICAgUEBAMEBgUGBgYFBgYGBwkIBgcJBwYGCAsICQo(...TRUNCATED) | 756 | 0 | null | What object is indicated by the polygon-marked area in this image? | What object is indicated by the polygon-marked area in this image? | instrumentrecognition | CholecSeg8k | CholecSeg8k_InstrumentRecognition_train_01.tsv | surgbench_d_norewrite | 0 |
Bipolar Forceps | Large Needle Driver | Clip Applier | Monopolar Curved Scissors | D | null | "[{\"source\": \"PDF\", \"id\": \"Principles of Laparoscopy- Trocar positioning and placement.pdf:c1(...TRUNCATED) | null | "{\"q_rewrite\": {\"question\": \"Which surgical instrument is illustrated in the following image?\"(...TRUNCATED) | "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAIBAQEBAQIBAQECAgICAgQDAgICAgUEBAMEBgUGBgYFBgYGBwkIBgcJBwYGCAsICQo(...TRUNCATED) | 532 | 0 | null | Which surgical instrument is illustrated in the following image? | Can you identify the instrument within the contour in this image? | instrumentrecognition | GraSP | GraSP_shorttermInstrumentRecognition_train_01.tsv | surgbench_d_norewrite | 1 |
cottle | haemostatic_foam | kerrisons | spatula_dissector | C | null | "[{\"source\": \"PDF\", \"id\": \"kupdf.com_a-complete-hospital-manual-of-instruments-and-procedures(...TRUNCATED) | null | "{\"q_rewrite\": {\"question\": \"Which surgical instrument is illustrated in the images below?\", \(...TRUNCATED) | "['/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAIBAQEBAQIBAQECAgICAgQDAgICAgUEBAMEBgUGBgYFBgYGBwkIBgcJBwYGCAsIC(...TRUNCATED) | 394 | 0 | null | Which surgical instrument is illustrated in the images below? | Identify the instrument in these images. | instrumentrecognition | pitvis | pitvis_instrumentRecognition_train_01.tsv | surgbench_d_norewrite | 1 |
Prograsp Forceps | Bipolar Forceps | Laparoscopic Grasper | Large Needle Driver | B | null | "[{\"source\": \"PDF\", \"id\": \"Surgical care at the district hospital.pdf:c303\", \"snippet\": \"(...TRUNCATED) | null | "{\"q_rewrite\": {\"question\": \"Which surgical instrument is depicted within the highlighted regio(...TRUNCATED) | "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAIBAQEBAQIBAQECAgICAgQDAgICAgUEBAMEBgUGBgYFBgYGBwkIBgcJBwYGCAsICQo(...TRUNCATED) | 563 | 0 | null | Which surgical instrument is depicted within the highlighted region of the diagram? | What instrument is indicated by the area within the bounding box in this image? | instrumentrecognition | PSI-AVA_Laparoscopic | PSI-AVA_InstrumentRecognition_train_01.tsv | surgbench_d_norewrite | 1 |
Prograsp Forceps | Clip Applier | Monopolar Curved Scissors | Suction Instrument | C | null | "[{\"source\": \"PDF\", \"id\": \"kupdf.com_a-complete-hospital-manual-of-instruments-and-procedures(...TRUNCATED) | null | "{\"q_rewrite\": {\"question\": \"Identify the surgical instrument that is highlighted within the bo(...TRUNCATED) | "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAIBAQEBAQIBAQECAgICAgQDAgICAgUEBAMEBgUGBgYFBgYGBwkIBgcJBwYGCAsICQo(...TRUNCATED) | 491 | 0 | null | "Identify the surgical instrument that is highlighted within the boxed area of the provided diagram.(...TRUNCATED) | What is the name of the instrument marked by the bounding box in this image? | instrumentrecognition | PSI-AVA_Laparoscopic | PSI-AVA_InstrumentRecognition_train_01.tsv | surgbench_d_norewrite | 1 |
Gastrointestinal Tract | Cystic Duct | Grasper | Gallbladder | C | null | "[{\"source\": \"PDF\", \"id\": \"Pancreatic Cyst-Gastrostomy .pdf:c16\", \"snippet\": \"tube intrao(...TRUNCATED) | null | "{\"q_rewrite\": {\"question\": \"Which surgical instrument is most likely represented by the polygo(...TRUNCATED) | "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAIBAQEBAQIBAQECAgICAgQDAgICAgUEBAMEBgUGBgYFBgYGBwkIBgcJBwYGCAsICQo(...TRUNCATED) | 531 | 0 | null | "Which surgical instrument is most likely represented by the polygon-marked area in the attached ima(...TRUNCATED) | What object is indicated by the polygon-marked area in this image? | instrumentrecognition | CholecSeg8k | CholecSeg8k_InstrumentRecognition_train_01.tsv | surgbench_d_norewrite | 1 |
Laparoscopic Grasper | Prograsp Forceps | Large Needle Driver | Suction Instrument | A | null | "[{\"source\": \"PDF\", \"id\": \"Laparoscopic-Right-Colectomy.pdf:c10\", \"snippet\": \"dot) stands(...TRUNCATED) | null | "{\"q_rewrite\": {\"question\": \"Which instrument is illustrated in the highlighted contour of the (...TRUNCATED) | "/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAIBAQEBAQIBAQECAgICAgQDAgICAgUEBAMEBgUGBgYFBgYGBwkIBgcJBwYGCAsICQo(...TRUNCATED) | 1,479 | 0 | null | Which instrument is illustrated in the highlighted contour of the provided laparoscopic image? | Identify the instrument within the contour in this image. | instrumentrecognition | GraSP | GraSP_shorttermInstrumentRecognition_train_01.tsv | surgbench_d_norewrite | 1 |
End of preview.
SurgBench CoT smoke-test data
A portable, 24-item smoke-test input for the SurgBench CoT generator/verifier pipeline.
Contents
data/test_smoke.jsonl: canonical train-format records with question, four options, gold index, and a relative image path.data/test_smoke_with_base64.jsonl: the same 24 records with frames embedded as base64 text, for transfer between clusters.
Usage
Download the Dataset. From its root, materialize the images with the companion code repository:
python /path/to/reasoning/scripts/materialize_base64_smoketest.py \
--input data/test_smoke_with_base64.jsonl \
--output data/test_smoke.jsonl \
--image-dir images
Then run the pipeline from the Dataset root. The canonical JSONL references images/smoke_000.jpg-style relative paths.
This is a small reproducibility fixture only. It is not a training split, benchmark, or redistribution of the complete SurgBench corpus.
- Downloads last month
- 88