world-model-data-benchmark-lab / evaluation_template.csv
Agenten's picture
Upload 8 files
9ecc04a verified
Raw History Blame Contribute Delete
305 Bytes
model,split,horizon,prediction_error,drift_ratio,action_fidelity,task_success,latency_ms,memory_mb
example-model,test,1,0.021,1.00,0.94,0.81,4.8,920
example-model,test,5,0.048,2.29,0.90,0.78,22.7,920
example-model,test,10,0.091,4.33,0.83,0.71,45.1,920
example-model,test,25,0.204,9.71,0.69,0.54,111.3,920