The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
Error code: DatasetGenerationError
Exception: IndexError
Message: list index out of range
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1848, in _prepare_split_single
original_shard_lengths[original_shard_id] += len(table)
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^
IndexError: list index out of range
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/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 1683, 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 1869, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
text string |
|---|
45 0.053474 0.832304 0.044843 0.077992 |
15 0.362378 0.864736 0.048751 0.067696 |
44 0.111564 0.696332 0.045277 0.078636 |
7 0.182410 0.899807 0.050380 0.074517 |
11 0.233279 0.937773 0.051574 0.075032 |
15 0.290174 0.972973 0.049837 0.054054 |
39 0.179533 0.735972 0.041585 0.074003 |
49 0.049294 0.667632 0.042128 0.075804 |
7 0.246851 0.774646 0.046580 0.074131 |
23 0.466558 0.959974 0.051031 0.071042 |
19 0.416504 0.907272 0.051249 0.072716 |
41 0.006840 0.639254 0.013681 0.062291 |
48 0.955320 0.924775 0.027036 0.036165 |
48 0.307106 0.924775 0.031882 0.036165 |
2 0.749077 0.240798 0.045168 0.075418 |
42 0.795603 0.197941 0.036374 0.075161 |
4 0.983985 0.053411 0.032030 0.073359 |
6 0.434745 0.710618 0.051683 0.062677 |
2 0.807220 0.380116 0.047231 0.075547 |
26 0.611944 0.395238 0.047557 0.071557 |
34 0.712324 0.276384 0.044191 0.056757 |
2 0.845385 0.337645 0.038436 0.075290 |
6 0.453800 0.664028 0.053746 0.063707 |
46 0.872910 0.130888 0.041802 0.078250 |
22 0.564332 0.472523 0.047231 0.068597 |
36 0.960369 0.240605 0.037134 0.071171 |
38 0.495277 0.908237 0.055483 0.050450 |
26 0.688599 0.524646 0.049946 0.070399 |
18 0.492182 0.588739 0.054506 0.063192 |
38 0.394463 0.828636 0.055157 0.052381 |
6 0.525570 0.819369 0.051792 0.059331 |
36 0.928122 0.093436 0.036048 0.070528 |
46 0.834148 0.168468 0.041802 0.074389 |
31 0.668838 0.331982 0.044517 0.071429 |
14 0.557709 0.732883 0.043105 0.058559 |
14 0.479207 0.617246 0.043540 0.060232 |
14 0.543865 0.776963 0.054506 0.059717 |
46 0.884799 0.286680 0.036048 0.038996 |
2 0.054180 0.240798 0.045168 0.075418 |
42 0.100706 0.197941 0.036374 0.075161 |
20 0.519544 0.100000 0.033008 0.073874 |
4 0.292128 0.053411 0.038111 0.073359 |
28 0.601683 0.075097 0.029859 0.071943 |
2 0.112324 0.380116 0.047231 0.075547 |
4 0.325624 0.203861 0.038219 0.073359 |
44 0.739739 0.037902 0.034311 0.075804 |
13 0.909826 0.029215 0.031379 0.058430 |
34 0.019761 0.276384 0.039522 0.056757 |
2 0.150489 0.337645 0.038436 0.075290 |
12 0.441911 0.137323 0.034093 0.070270 |
46 0.178013 0.130888 0.041802 0.078250 |
36 0.265472 0.240605 0.037134 0.071171 |
44 0.701629 0.046654 0.031488 0.072973 |
36 0.233225 0.093436 0.036048 0.070528 |
46 0.139251 0.168468 0.041802 0.074389 |
8 0.408469 0.156242 0.037568 0.071815 |
24 0.571118 0.083591 0.028882 0.072458 |
5 0.839142 0.031789 0.032682 0.061776 |
36 0.789794 0.033526 0.031922 0.067053 |
46 0.189902 0.286680 0.036048 0.038996 |
45 0.748371 0.137323 0.044843 0.077992 |
7 0.877307 0.204826 0.050380 0.074517 |
29 0.481705 0.043501 0.037459 0.077992 |
11 0.928176 0.242793 0.051574 0.075032 |
15 0.980076 0.282561 0.039848 0.063192 |
44 0.044843 0.037902 0.034311 0.075804 |
49 0.691260 0.107465 0.042020 0.076448 |
13 0.214929 0.029215 0.031379 0.058430 |
39 0.874430 0.040991 0.041585 0.074003 |
7 0.941748 0.079665 0.046580 0.074131 |
44 0.011238 0.046654 0.022476 0.072973 |
25 0.421498 0.035457 0.037351 0.070142 |
41 0.638219 0.085650 0.042780 0.076577 |
1 0.604343 0.079086 0.043865 0.079665 |
21 0.364658 0.032497 0.035071 0.064994 |
5 0.144245 0.031789 0.032682 0.061776 |
33 0.551629 0.060682 0.040065 0.077864 |
36 0.094897 0.033526 0.031922 0.067053 |
47 0.679913 0.859073 0.060586 0.065380 |
47 0.779425 0.779408 0.059826 0.063578 |
45 0.053474 0.137323 0.044843 0.077992 |
15 0.362378 0.169755 0.048751 0.067696 |
47 0.635722 0.762548 0.061889 0.061776 |
47 0.735071 0.681918 0.059718 0.061647 |
7 0.182410 0.204826 0.050380 0.074517 |
11 0.233279 0.242793 0.051574 0.075032 |
15 0.290174 0.282561 0.049837 0.063192 |
40 0.532519 0.561326 0.060695 0.064479 |
39 0.179533 0.040991 0.041585 0.074003 |
7 0.246851 0.079665 0.046580 0.074131 |
23 0.386156 0.375032 0.051792 0.065894 |
19 0.337296 0.327928 0.054615 0.071557 |
3 0.625407 0.464929 0.059501 0.069112 |
23 0.466558 0.264994 0.051031 0.071042 |
19 0.416504 0.212291 0.051249 0.072716 |
51 0.593920 0.663578 0.057980 0.060489 |
27 0.434799 0.433076 0.058523 0.065637 |
27 0.521010 0.323810 0.055266 0.070785 |
31 0.479316 0.485907 0.054397 0.062934 |
48 0.955320 0.229794 0.027036 0.036165 |
lgd-cards-video-day3 — day-3 deck-spread card-pip tiles
Training data for the Live Game Defender card detector, day-3
campaign (2026-07-20, KAS-52). This is the set that trained
sroot/lgd-cards-gen4 — the spread-recall fine-tune.
⚠️ Not casino ground truth. Our own PoC / Czech Croupier Academy footage. Labels are detector-proposed + OpenAI-
gpt-5-mini-verified against the closed 52-code vocabulary — LLM- verified, not human casino annotation (rule #5). CC-BY-NC-4.0.
What's here
YOLO detection tiles of playing-card corner index pips (52 classes AS, 10H, KD, …), in the
serve-time tiling geometry (a 4K frame → 6×4 overlap tiles; empty tiles kept as felt negatives).
| Split | Images | Labels |
|---|---|---|
images/train + labels/train |
3,832 | 3,832 |
images/val + labels/val |
630 | 630 |
Plus boxes.jsonl (per-box provenance: detector code + confidence, OpenAI code, agree flag) and
manifest.json (run config + counts). Internal QA review/ crops are excluded from this upload.
Source & method
From 52_cards.mp4 — a 71-minute 4K overhead recording of a dealer repeatedly spreading the full
52-card deck face-up. Because most of the video is shuffle/deal/empty, the deck-spread windows were
detected and the video trimmed to ~2,650 s of spread footage before labeling, then sampled at
0.15 fps. autolabel_video.py proposed pip boxes with the served detector (floor 0.10) and had
gpt-5-mini (reasoning minimal) verify/name each against the 52-code vocabulary. 490 frames →
16,875 verified boxes for $3.74. The held-out spread video went to
lgd-cards-holdout, never trained on.
Family
Datasets: day1 · day2 · day3 (this) · holdout · Model trained: lgd-cards-gen4. The third-party Roboflow ow27d base set is not redistributed.
- Downloads last month
- 1,415