SolarRenderer: Few-Step Generative Reference Rendering
Python
docs: eval README rates the causal variant from its released renders and verdicts
Enderfgaaaef2f7ed8ed
diff --git a/eval/README.md b/eval/README.mdindex 57694f8..33d98df 100644--- a/eval/README.md+++ b/eval/README.md@@ -73,7 +73,8 @@ python eval/summary.py --scores ./scores --methods solarrenderer,other --split hpython eval/summary.py --scores ./scores --methods solarrenderer,other --split heldout```-On the released renders (`results/solarrenderer`) these commands give the paper's means; DOVER and VideoScore2 depend+On the released renders (`results/solarrenderer`, and `results/solarrenderer_causal` for the causal variant) these+commands give the paper's means; DOVER and VideoScore2 dependslightly on library versions (VideoScore2 generates its answer, so expect differences of about 0.02).Results are one JSON per case and metric, under `<out>/<metric>/<method>/hard/<case_id>.json` and@@ -144,3 +145,13 @@ python eval/elo.py ./SolarRenderer-Data/results/judge/hard/*.jsonl --split hardcontrol_model_50step,official_backbone_50step,pai_controlnet_union,streaming_cwm,streamrender_h3,cosmos_transfer2_5,\wan2_2_vace_fun```++The causal variant (`results/solarrenderer_causal`) is judged only against SolarRenderer. Adding its verdicts and+`solarrenderer_causal` as a tenth method rates it and moves the other nine by a few points at most; shifting all ten+ratings so that the other nine average 1000 gives its row in the paper:++```bash+python eval/elo.py ./SolarRenderer-Data/results/judge/hard/*.jsonl --split hard --methods solarrenderer,lynnreal_omni,\+control_model_50step,official_backbone_50step,pai_controlnet_union,streaming_cwm,streamrender_h3,cosmos_transfer2_5,\+wan2_2_vace_fun,solarrenderer_causal+```