has now been downloaded over 5,400 times. I am a…
0 reactions
has now been downloaded over 5,400 times.
I am a figurative painter. I am not a developer. I do not know what most of you are doing with it, and I would genuinely like to know.
For those who missed the first post: roughly 3,000 to 4,000 documented works, the human figure as sustained subject across five decades, oil on canvas, works on paper, drawings, etchings, lithographs, and digital works. CC-BY-NC-4.0, artist-controlled, full provenance metadata. My total output is approximately double what is currently published and I am adding to it continuously. It is a living record, not a monument.
**If you fine-tune on it** — post the results. I want to see what fifty years of a single figurative practice produces when a model trains on it.
**If you are a researcher** — the dataset is citable. It is one of the few fine art datasets of this scale that is properly licensed, published with artist consent, and carries full metadata.
**If you find errors in the metadata** — please flag them. I built this myself. Title, date, and medium corrections are welcome.
Dataset: huggingface.co/datasets/Hafftka/michael-hafftka-catalog-raisonne http://huggingface.co/datasets/Hafftka/michael-hafftka-catalog-raisonne
#technology
source 5,400 downloads later — what are you doing with my catalog raisonné?
A few weeks ago I posted that I had published my catalog raisonné as an open dataset on Hugging Face. It
I am a figurative painter. I am not a developer. I do not know what most of you are doing with it, and I would genuinely like to know.
For those who missed the first post: roughly 3,000 to 4,000 documented works, the human figure as sustained subject across five decades, oil on canvas, works on paper, drawings, etchings, lithographs, and digital works. CC-BY-NC-4.0, artist-controlled, full provenance metadata. My total output is approximately double what is currently published and I am adding to it continuously. It is a living record, not a monument.
**If you fine-tune on it** — post the results. I want to see what fifty years of a single figurative practice produces when a model trains on it.
**If you are a researcher** — the dataset is citable. It is one of the few fine art datasets of this scale that is properly licensed, published with artist consent, and carries full metadata.
**If you find errors in the metadata** — please flag them. I built this myself. Title, date, and medium corrections are welcome.
Dataset: huggingface.co/datasets/Hafftka/michael-hafftka-catalog-raisonne http://huggingface.co/datasets/Hafftka/michael-hafftka-catalog-raisonne
#technology
source 5,400 downloads later — what are you doing with my catalog raisonné?
A few weeks ago I posted that I had published my catalog raisonné as an open dataset on Hugging Face. It
| 3,000 mlx | |
| 4 | |
| 0 |