Support grad for depth only in rasterizer

[github-actions[bot]]

Running experiment on sagemaker with git sha 7e2e1f2d2f819a6908941292549e771dac3f21da

[github-actions[bot]]

Training job grad-depth-7e2e1f2-230909-005041-tat-truck-baseline created

[github-actions[bot]]

Running experiment on sagemaker with git sha 7e2e1f2d2f819a6908941292549e771dac3f21da

[github-actions[bot]]

Training job grad-depth-7e2e1f2-230909-005149-tat-train-baseline created

[github-actions[bot]]

Training job grad-depth-7e2e1f2-230909-005041-tat-truck-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/grad-depth-7e2e1f2-230909-005041-tat-truck-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/grad-depth-7e2e1f2-230909-005041-tat-truck-baseline/output/output.tar.gz

[github-actions[bot]]

Training job grad-depth-7e2e1f2-230909-005041-tat-truck-baseline final metrics:

Latest Metrics

train:iteration	train:l1loss	train:loss	train:num_valid_points	train:psnr	train:ssim	train:ssimloss	val:loss	val:psnr	val:ssim
30000.0	0.02365143597126007	0.04058298096060753	442217.0	27.55559730529785	0.8916908502578735	0.10830914974212646	0.05380209535360336	25.163917541503906	0.8650290369987488

Max Metrics

train:5kpsnr	train:5kssim	train:7kpsnr	train:7kssim	train:psnr	train:ssim	val:5kpsnr	val:5kssim	val:7kpsnr	val:7kssim	val:psnr	val:ssim
24.77243423461914	0.8409563302993774	24.086151123046875	0.8653521537780762	27.55559730529785	0.8916908502578735	23.030521392822266	0.8133056163787842	23.853111267089844	0.8373969197273254	25.163917541503906	0.8650290369987488

[github-actions[bot]]

Training job grad-depth-7e2e1f2-230909-005149-tat-train-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/grad-depth-7e2e1f2-230909-005149-tat-train-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/grad-depth-7e2e1f2-230909-005149-tat-train-baseline/output/output.tar.gz

[github-actions[bot]]

Training job grad-depth-7e2e1f2-230909-005149-tat-train-baseline final metrics:

Latest Metrics

train:iteration	train:l1loss	train:loss	train:num_valid_points	train:psnr	train:ssim	train:ssimloss	val:loss	val:psnr	val:ssim
30000.0	0.027597062289714813	0.041713614016771317	677702.0	26.506502151489258	0.901820182800293	0.09817981719970703	0.09981270134449005	20.97577476501465	0.801036536693573

Max Metrics

train:5kpsnr	train:5kssim	train:7kpsnr	train:7kssim	train:psnr	train:ssim	val:5kpsnr	val:5kssim	val:7kpsnr	val:7kssim	val:psnr	val:ssim
18.4986572265625	0.740445613861084	19.621097564697266	0.780716061592102	26.506502151489258	0.901820182800293	18.968416213989258	0.7297627329826355	19.654048919677734	0.7564723491668701	20.97577476501465	0.801036536693573

[github-actions[bot]]

Running experiment on sagemaker with git sha 85724d7bc0632f3ab7a66b0c8cd01f0b4f98bb57

[github-actions[bot]]

Training job grad-depth-85724d7-230910-044828-tat-truck-baseline created

[github-actions[bot]]

Running experiment on sagemaker with git sha 85724d7bc0632f3ab7a66b0c8cd01f0b4f98bb57

[github-actions[bot]]

Training job grad-depth-85724d7-230910-044944-tat-train-baseline created

[github-actions[bot]]

Training job grad-depth-85724d7-230910-044828-tat-truck-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/grad-depth-85724d7-230910-044828-tat-truck-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-truck-baseline/grad-depth-85724d7-230910-044828-tat-truck-baseline/output/output.tar.gz

[github-actions[bot]]

Training job grad-depth-85724d7-230910-044828-tat-truck-baseline final metrics:

Latest Metrics

train:iteration	train:l1loss	train:loss	train:num_valid_points	train:psnr	train:ssim	train:ssimloss	val:loss	val:psnr	val:ssim
30000.0	0.035545192658901215	0.059377893805503845	439447.0	23.571882247924805	0.8452913165092468	0.15470868349075317	0.05419863015413284	25.105846405029297	0.8648907542228699

Max Metrics

train:5kpsnr	train:5kssim	train:7kpsnr	train:7kssim	train:psnr	train:ssim	val:5kpsnr	val:5kssim	val:7kpsnr	val:7kssim	val:psnr	val:ssim
22.985036849975586	0.81522536277771	23.676334381103516	0.8425229787826538	23.571882247924805	0.8452913165092468	23.084001541137695	0.8149509429931641	23.93067169189453	0.838331937789917	25.105846405029297	0.8648907542228699

[github-actions[bot]]

Training job grad-depth-85724d7-230910-044944-tat-train-baseline completed.

Model url: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/grad-depth-85724d7-230910-044944-tat-train-baseline/output/model.tar.gz,

tensorboard output path: s3://taichi-3d-gaussian-splatting-log/tat-train-baseline/grad-depth-85724d7-230910-044944-tat-train-baseline/output/output.tar.gz

[github-actions[bot]]

Training job grad-depth-85724d7-230910-044944-tat-train-baseline final metrics:

Latest Metrics

train:iteration	train:l1loss	train:loss	train:num_valid_points	train:psnr	train:ssim	train:ssimloss	val:loss	val:psnr	val:ssim
30000.0	0.04131251946091652	0.06635285913944244	668398.0	23.763450622558594	0.8334857821464539	0.16651421785354614	0.09747117757797241	21.061948776245117	0.8045176267623901

Max Metrics

train:5kpsnr	train:5kssim	train:7kpsnr	train:7kssim	train:psnr	train:ssim	val:5kpsnr	val:5kssim	val:7kpsnr	val:7kssim	val:psnr	val:ssim
22.968643188476562	0.8369643688201904	22.771406173706055	0.8545178174972534	23.763450622558594	0.8334857821464539	19.088722229003906	0.731535017490387	19.672351837158203	0.7568982243537903	21.061948776245117	0.8045176267623901