datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
bokeh-eval-metricasbokeh-eval-lfrepro-nossobokeh-eval-lfrepro-rotacbokeh-eq4-lfrepro-nossobokeh-eq4-lfrepro-semtreinobokeh-eval-lfrepro-oficialbokeh-eval-lfrepro-kfixbokeh-eval-lfrepro-rotac45kbokeh-eval-lfrepro-baseA15k-fullbokeh-eval-lfrepro-nofilter28kbokeh-eval-lfrepro-rotac60kbokeh-eval-lfrepro-semtreinobokeh-eq4-lfrepro-oficialbokeh-eq4-lfrepro-rotac60kbokeh-eq4-lfrepro-fase1bokeh-kesc5-ebb-nossobokeh-kesc5-ebb-oficialbokeh-eq4-ebb-oficialbokeh-eval-lfrepro-geoB15k-fullbokeh-eq4-ebb-kfixbokeh-kesc1-ebb-oficialbokeh-eval-lfrepro-fase1bokeh-eq4-ebb-fase1bokeh-kesc1-ebb-nossobokeh-eq4-ebb-semtreinobokeh-eq4-ebb-nossobokeh-eq4-ebb-nofilterbokeh-eval-official
bokeh-eval-official
Bokeh synthesis artifacts: all-in-focus input, ground-truth bokeh, the
best-K render and the full K sweep, one split per benchmark.
Generated by inference/bokeh_net.py.
bokeh-eq4-lfrepro-nofilterbokeh-eq4-rb-oficial