Unable to reproduce results on Mip-NeRF 360 Captures

[animesh-77]

As mentioned here https://docs.gsplat.studio/main/tests/eval.html#reproduced-metrics , after installing gsplat library I ran benchmarks/basic.sh which uses simple_trainer.py to train on Mip-NeRF 360 Captures.

For e.g. the bicycle scene was evaluated with DATA_FACTOR 4 and the Inria implementation resulted in ≈6M splats. However running benchmarks/basic.sh results in a file much smaller than this with ≈700K splats ONLY!

However, training with ns-train splatfacto results in results similar to Inria with ≈6M splats. Can anyone please help me understand what I am doing wrong here? Thanks!