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 7 new columns ({'index_in_numpy_array', 'filename', 'id', 'text_token_length', 'vq_token_length', 'iconshop_description', 'label'}) and 1 missing columns ({'simplified_svg_file_path'}).
This happened while the csv dataset builder was generating data using
hf://datasets/Potpov/grimoire-figr8/tokenized/split.csv (at revision 5b14417f7c2cc28930820cc268aa94aa413e2502), ['hf://datasets/Potpov/grimoire-figr8@5b14417f7c2cc28930820cc268aa94aa413e2502/figr8_vsq_split.csv', 'hf://datasets/Potpov/grimoire-figr8@5b14417f7c2cc28930820cc268aa94aa413e2502/tokenized/split.csv']
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 1837, 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 2369, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2297, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
index_in_numpy_array: int64
class: string
split: string
filename: string
iconshop_description: string
id: string
vq_token_length: int64
text_token_length: int64
description: string
label: string
-- schema metadata --
pandas: '{"index_columns": [{"kind": "range", "name": null, "start": 0, "' + 1488
to
{'simplified_svg_file_path': Value('string'), 'class': Value('string'), 'description': Value('string'), 'split': Value('string')}
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 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 1839, 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 7 new columns ({'index_in_numpy_array', 'filename', 'id', 'text_token_length', 'vq_token_length', 'iconshop_description', 'label'}) and 1 missing columns ({'simplified_svg_file_path'}).
This happened while the csv dataset builder was generating data using
hf://datasets/Potpov/grimoire-figr8/tokenized/split.csv (at revision 5b14417f7c2cc28930820cc268aa94aa413e2502), ['hf://datasets/Potpov/grimoire-figr8@5b14417f7c2cc28930820cc268aa94aa413e2502/figr8_vsq_split.csv', 'hf://datasets/Potpov/grimoire-figr8@5b14417f7c2cc28930820cc268aa94aa413e2502/tokenized/split.csv']
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.
simplified_svg_file_path string | class string | description string | split string |
|---|---|---|---|
svgs_simplified/1000017-200.svg | fast forward | The audio player has fast-forward control for music and video. | train |
svgs_simplified/1000018-200.svg | next | Control the next music or video with the audio player. | train |
svgs_simplified/1000025-200.svg | record | Control audio player record music video. | train |
svgs_simplified/1000027-200.svg | stop | Stop music player audio. | train |
svgs_simplified/1000028-200.svg | man in love | Man in love has an affair with passion in a romantic relationship on Valentine's day. | train |
svgs_simplified/1000029-200.svg | man in love | The man is in a romantic relationship filled with passion and is celebrating Valentine's Day. | train |
svgs_simplified/1000034-200.svg | man with hearts | The man had a romantic affair filled with passion and hearts, and it happened during Valentine's Day in their relationship. | train |
svgs_simplified/1000038-200.svg | man in love | Man in love has a passionate affair in a romantic relationship on Valentine's Day. | train |
svgs_simplified/1000039-200.svg | man with hearts | Man with hearts had a romantic Valentine affair with passion in their relationship. | train |
svgs_simplified/1000040-200.svg | man with hearts | Man has affair with romantic passion on Valentine's Day. | train |
svgs_simplified/1000043-200.svg | man refresh | Man wants to refresh. | train |
svgs_simplified/1000044-200.svg | man in love | The man-in-love had an affair full of passion within his romantic relationship during Valentine's Day. | train |
svgs_simplified/1000051-200.svg | man refresh | The man rejuvenated himself. | train |
svgs_simplified/1000055-200.svg | couple | The couple married. | train |
svgs_simplified/1000058-200.svg | couple in love | The couple in love got married. | train |
svgs_simplified/1000059-200.svg | couple in love | The couple in love got married. | train |
svgs_simplified/1000083-200.svg | map pin | Caution: Map pin with pin notice warning. | train |
svgs_simplified/1000087-200.svg | mine | Mine exploded. | train |
svgs_simplified/100009-200.svg | sunny | The forecast for today is partly cloudy with some sun. | train |
svgs_simplified/1000098-200.svg | cloud | The server is on the cloud. | train |
svgs_simplified/1000099-200.svg | flag | The flag has the emblem as its mark. | train |
svgs_simplified/1000102-200.svg | flag | The flag has a mark. | train |
svgs_simplified/1000104-200.svg | book mark | Bookmark the page with a flag. | train |
svgs_simplified/1000111-200.svg | delete file | Delete file. | train |
svgs_simplified/1000115-200.svg | funnel | Sort the arrayed data to refine the funnel and filter in order. | train |
svgs_simplified/1000121-200.svg | add file | Add new text by writing an article with additional information. | train |
svgs_simplified/1000128-200.svg | business man | A businessman works with people. | train |
svgs_simplified/1000130-200.svg | advertising | Advertising news with a bullhorn or megaphone. | train |
svgs_simplified/1000132-200.svg | lecture | The lecturer stood behind the podium at the speech desk during the presentation. | train |
svgs_simplified/1000133-200.svg | business idea | The businessman had a business idea during the brainstorming session when a light bulb went off. | test |
svgs_simplified/1000137-200.svg | stategy | The businessman created a solution through strategy and planning. | train |
svgs_simplified/1000144-200.svg | snow flake | Snowflake is cold in winter on ice and snow. | train |
svgs_simplified/100015-200.svg | freezing rain | The daytime weather was freezing rain and ice. | train |
svgs_simplified/1000152-200.svg | snow flake | Snowflakes are hexagonal ice crystals that fall during winter. | train |
svgs_simplified/1000164-200.svg | add tag | Add tag. | train |
svgs_simplified/100017-200.svg | freezing rain | The forecast predicts freezing rain and icy weather on a cloudy daytime. | train |
svgs_simplified/1000171-200.svg | grid layout | Grid design is a layout. | train |
svgs_simplified/1000173-200.svg | design | Design an app layout and website design. | train |
svgs_simplified/1000190-200.svg | cleaver | The cleaver is a butcher knife. | train |
svgs_simplified/1000194-200.svg | knife | Use the knife blade to cut with cutlery. | train |
svgs_simplified/1000196-200.svg | knife | Use the knife blade to cut with cutlery. | train |
svgs_simplified/1000200-200.svg | crosshair | Aim at the target with the crosshair. | train |
svgs_simplified/1000201-200.svg | globe | Earth is a planet. | train |
svgs_simplified/1000203-200.svg | smart phone | Mobile phone. | train |
svgs_simplified/1000206-200.svg | alarm clock | The alarm clock sounded at the scheduled time. | train |
svgs_simplified/1000208-200.svg | atom | Atom is studied in nuclear and chemistry in physics. | train |
svgs_simplified/1000210-200.svg | briecase | I picked up my bag. | train |
svgs_simplified/1000211-200.svg | puzzle piece | Game toy has puzzle piece. | test |
svgs_simplified/1000213-200.svg | question | What is a FAQ and why is there a question mark inside a circle? | train |
svgs_simplified/1000223-200.svg | list | To do list. | test |
svgs_simplified/1000225-200.svg | share | Share content on social media by sharing. | train |
svgs_simplified/1000227-200.svg | share | Sharing social content. | train |
svgs_simplified/1000229-200.svg | lock | The lock is private. | test |
svgs_simplified/1000235-200.svg | unlock | The padlock unlocked. | val |
svgs_simplified/1000239-200.svg | unlock | Unlock the padlock. | train |
svgs_simplified/1000240-200.svg | file download | Download the file with the arrow pointing down. | train |
svgs_simplified/1000248-200.svg | file zip | File is compressed as rar. | test |
svgs_simplified/1000250-200.svg | file text | The file contains text in a document format. | train |
svgs_simplified/1000264-200.svg | file text | The file is a txt format. | train |
svgs_simplified/1000295-200.svg | repeat | Repeat arrows. | train |
svgs_simplified/1000298-200.svg | pi chart | An infographic displays a pie chart. | train |
svgs_simplified/1000299-200.svg | scan | Scan with the document scanner. | train |
svgs_simplified/1000300-200.svg | printer | Printer prints. | train |
svgs_simplified/1000303-200.svg | monitor | Screen monitor. | train |
svgs_simplified/1000316-200.svg | news paper | Newspaper. | train |
svgs_simplified/1000319-200.svg | link | Link the chain. | train |
svgs_simplified/1000320-200.svg | target | Target. | train |
svgs_simplified/1000329-200.svg | star | Favorite star. | train |
svgs_simplified/1000331-200.svg | image | Image/picture. | train |
svgs_simplified/1000333-200.svg | stop watch | Stopwatch timer. | train |
svgs_simplified/1000338-200.svg | umbella | Umbrella. | train |
svgs_simplified/1000342-200.svg | television | TV screen. | train |
svgs_simplified/1000343-200.svg | conflict | Arrows clashed with an impact in the conflict. | train |
svgs_simplified/1000344-200.svg | open | Open the up-arrow. | train |
svgs_simplified/1000347-200.svg | battery | Battery low. | train |
svgs_simplified/1000348-200.svg | pi chart | A pie chart is an infographic representation. | test |
svgs_simplified/1000350-200.svg | sign | Keep left on the street. | train |
svgs_simplified/1000356-200.svg | sign | Attention! Street sign. Stop. Warning. | train |
svgs_simplified/1000359-200.svg | sign | Sign warns to stop at street with attention to road. | train |
svgs_simplified/1000360-200.svg | sign | This street has a curve and a turn. | train |
svgs_simplified/1000362-200.svg | sign | Sign of caution ahead at the junction of narrow road and street. | train |
svgs_simplified/1000367-200.svg | traffic light | The traffic light controls the junction at the crossroads. | train |
svgs_simplified/1000368-200.svg | star | Stars shine in the sky. | test |
svgs_simplified/1000369-200.svg | star | Stars shine in the night sky. | train |
svgs_simplified/1000372-200.svg | star | Stars shine in the sky. | train |
svgs_simplified/1000374-200.svg | star | Star. | train |
svgs_simplified/1000375-200.svg | star | Stars shine in the sky. | train |
svgs_simplified/1000430-200.svg | car | The car has an alarm signal timer. | test |
svgs_simplified/1000434-200.svg | car | The car used internet cloud storage for vehicle data. | train |
svgs_simplified/1000443-200.svg | car | An email message was sent in an envelope about a car automobile vehicle. | train |
svgs_simplified/1000448-200.svg | car | The car's image was captured by a camera. | train |
svgs_simplified/1000455-200.svg | car | Resend the vehicle. | train |
svgs_simplified/1000456-200.svg | car | The car went forward and then returned to repeat the movement. | train |
svgs_simplified/1000458-200.svg | car | My favorite bookmark is a star-shaped one with a car on it. | train |
svgs_simplified/1000467-200.svg | car | The car's temperature in degrees was high. | train |
svgs_simplified/1000469-200.svg | car | Car error: vehicle rejected, unavailable. | val |
svgs_simplified/1000474-200.svg | motor cycle | The motorcycle is a vehicle for transport at the cafe. | train |
svgs_simplified/1000475-200.svg | boat | Transport the boat with the car using the trailer. | test |
svgs_simplified/1000479-200.svg | arrow | Right arrow. | test |
svgs_simplified/1000480-200.svg | arrow | Right arrow. | train |
End of preview.
Grimoire — FIGR-8 (processed / final training data)
The final, post-processed FIGR-8 training data for the Grimoire SVG pipeline. Shipped processed because regenerating it from the raw FIGR-8 corpus (convert + deepsvg normalize + tokenize) is painful.
⚠️ Split differs from the paper. The paper describes a 427K-sample, 1,000-class, 90/5/5 subset (that split file is
datasets/make_font/figr8_final_paper_split.csvin the code repo), but it is not reproducible from code (the class selection / spelling fixes / seed are not in the repo). The split shipped here — and used to train the released checkpoints — is 373,388 samples, 16,486 classes, 80/10/10 (convert_figr8.py, per-classtrain_test_split(test_size=0.1, random_state=42)).
Contents
svgs_simplified.tar.gz— the deepsvg-normalized SVG corpus (374,992*.svg), the input for stage-1 (VSQ) training.figr8_vsq_split.csv— ready-to-use VSQ split (columnssimplified_svg_file_path,class,description,split); paths are relative to the extractedsvgs_simplified/. 99.6% of SVGs carry their real class + description.tokenized/tokenized.npy— precomputed VSQ token sequences for stage-2 (ART) training.tokenized/split.csv— train/val/test split + text; rows index intotokenized.npyviaindex_in_numpy_array.
Grimoire dataset family — recap
| dataset | what's pushed | plug-and-play for VSQ (stage 1)? | size |
|---|---|---|---|
| FIGR-8 (this repo) | processed SVGs + VSQ csv + ART tokens | ✅ yes — tar xzf svgs_simplified.tar.gz, then train on figr8_vsq_split.csv |
svgs 345 MB · csv 35 MB · tokens 530 MB |
| MNIST | raw PNGs + ART tokens | ⚠️ needs 1 easy step (PNG→tensors, create_precomputed_mnist.py) |
png 15 MB · tokens 3.7 GB |
| Emoji | processed SAM masks | ✅ yes — extract, CartoonDataset reads them directly |
5.7 GB |
| Cartoons | — not pushed (no experiments) | — | — |
Usage
- Stage 1 (VSQ) — plug and play:
tar xzf svgs_simplified.tar.gzand train onfigr8_vsq_split.csv(dataset keyfigr8, mlp decoder,num_segments=4,num_codes_per_shape=2,lseg=5.0,alpha=0.1). - Stage 2 (ART) — ready to train on
tokenized/{split.csv,tokenized.npy}with the stage-1 VSQ checkpoint from Potpov/grimoire-checkpointsfigr8/vsq.
Re-tokenize with scripts/tokenize_svg_dataset.py if you retrain the VSQ.
- Downloads last month
- 32